1 2011-01-31 00:00:03 <gavinandresen> I just know not a lot of people are using it to access the faucet.
  2 2011-01-31 00:00:21 <luke-jr> gavinandresen: /48 is the standard subnet size ISPs are supposed to give end users
  3 2011-01-31 00:00:30 <ArtForzZz> I love IPv6
  4 2011-01-31 00:00:36 <gavinandresen> (and implementing "rate limit based on first byte" worked nicely)
  5 2011-01-31 00:00:55 <echelon> the bitcoin times needs an editor :P
  6 2011-01-31 00:01:01 <luke-jr> basically the entire global IPv6 space is all under 2001::/16 and 2002::/16
  7 2011-01-31 00:01:13 <echelon> how old is that guy anyway
  8 2011-01-31 00:01:13 <jgarzik> gah.  bitcoin handles P2P nodes sanely, with select....  but then we have a separate thread that sits in an infinite loop, polling all connections for new P2P messages
  9 2011-01-31 00:01:19 <luke-jr> 2002:aabb:ccdd::/48 are automatically routed to the IPv4 aaa.bbb.ccc.ddd
 10 2011-01-31 00:01:25 <jgarzik> when it could simply process messages asynchronously
 11 2011-01-31 00:01:38 <ArtForzZz> really?
 12 2011-01-31 00:02:48 <jgarzik> ThreadMessageHandler2() is a loop that polls each datastream for new data, which is a separate thread from ThreadSocketHandler(), which is a loop that selects + reads/writes data
 13 2011-01-31 00:03:02 <ArtForzZz> hmmm... you're right. slightly weird design choice.
 14 2011-01-31 00:03:41 <ArtForzZz> conceptually it kinda makes sense somewhat
 15 2011-01-31 00:05:06 <jgarzik> A separate execution thread can make sense in some environments.  Polling each of N connections for new work, for each loop iteration, is strange.
 16 2011-01-31 00:05:20 <ArtForzZz> that is true
 17 2011-01-31 00:05:41 <jgarzik> Usually your I/O thread would queue one or more units of work to the execution thread.  The execution thread shouldn't be looping over each possible event source.
 18 2011-01-31 00:05:58 <ArtForzZz> yep
 19 2011-01-31 00:06:07 <ArtForzZz> exactly what I was going to say
 20 2011-01-31 00:06:51 <ArtForzZz> have the network thread push workitems of "got new message from node X" to the message handling thread
 21 2011-01-31 00:06:59 <jgarzik> yep
 22 2011-01-31 00:08:35 <ArtForzZz> with something like a Condition for "received messages is (non)empty"
 23 2011-01-31 00:08:56 <jgarzik> RPC user B stalls, waiting for RPC user A's HTTP data transmission.  Unlike P2P, which uses select and properly buffers.
 24 2011-01-31 00:09:01 <ArtForzZz> yep
 25 2011-01-31 00:09:18 <ArtForzZz> RPC I/O handling should be seperated from the core RPC stuff
 26 2011-01-31 00:10:23 <jgarzik> surely boost has an easy, epoll-like mechanism where you register event sources, and receive callbacks
 27 2011-01-31 00:10:49 <ArtForzZz> I'd hope so, I think any sane language/framework/whatever has
 28 2011-01-31 00:11:50 <ArtForzZz> with good async event support you don't really need any threading for stuff like this
 29 2011-01-31 00:15:57 <jgarzik> indeed
 30 2011-01-31 00:16:58 <gavinandresen> Well, bitcoin needed threading for mining on multi-cpus, so maybe Satoshi decided multithreading everything made sense.
 31 2011-01-31 00:17:15 <gavinandresen> Woulda coulda shoulda.....
 32 2011-01-31 00:17:28 <jgarzik> I can see event handling + network I/O buffering in a separate thread from execution, as execution covers database I/O and associated fsync's
 33 2011-01-31 00:17:43 <ArtForzZz> why?
 34 2011-01-31 00:17:51 <ArtForzZz> the Os already handles low level buffering for us
 35 2011-01-31 00:18:42 <jgarzik> our perf teams show it helps in some apps.
 36 2011-01-31 00:18:45 <ArtForzZz> and it's not like we are doing anything useful with those network messages while our main message handler thread is busy doing it's thing
 37 2011-01-31 00:19:05 <andrew12> ArtForzZz must be really sleepy
 38 2011-01-31 00:19:10 <ArtForzZz> yup
 39 2011-01-31 00:19:14 <jgarzik> definitely helped with latency when using those design choices at a $BigDatabaseVendor
 40 2011-01-31 00:19:19 <ArtForzZz> it's past 2AM here
 41 2011-01-31 00:19:26 <ArtForzZz> we're just queueing them up for the main handler
 42 2011-01-31 00:19:45 <ArtForzZz> it would make sense if were actually *processing* stuff in parallel, but we aren't
 43 2011-01-31 00:19:46 <jgarzik> but I write my own programs as single-thread async, FWIW
 44 2011-01-31 00:19:53 <jgarzik> life is easier
 45 2011-01-31 00:20:06 <andrew12> "Consecutive Vowels"
 46 2011-01-31 00:20:19 <ArtForzZz> we have threadreceiver handing things to threadhandler (which takes locks and sits there waiting for syncs) handing things to threadsender
 47 2011-01-31 00:21:02 <ArtForzZz> = we're really doing everything serialized anyway, only thing thrreading is doing here is to make the conectpual seperation more obvious
 48 2011-01-31 00:21:16 <jgarzik> if we delete CPU mining from official bitcoin, we could start unwinding some of this mess
 49 2011-01-31 00:21:29 <jgarzik> just leave getwork and push-mining
 50 2011-01-31 00:21:39 <ArtForzZz> you might as well have a single loop of (receive stuff - process stuff - send stuff)
 51 2011-01-31 00:21:52 <gavinandresen> jgarzik: I'm getting really tempted to do a major refactor and start creating a libbitcoin
 52 2011-01-31 00:21:56 <jgarzik> send is async
 53 2011-01-31 00:22:01 <jgarzik> rx + process
 54 2011-01-31 00:22:07 <ArtForzZz> really
 55 2011-01-31 00:22:14 <gavinandresen> jgarzik:  my heart wants to do it, but my head knows it's probably the wrong thing to do right now.
 56 2011-01-31 00:22:29 <ArtForzZz> I thought we also had a push_bytes_to_network thread
 57 2011-01-31 00:22:29 <jgarzik> gavinandresen: I hear ya :)
 58 2011-01-31 00:22:56 <jgarzik> ArtForzZz: ThreadSocketHandler for that, today, yes
 59 2011-01-31 00:23:38 <ArtForzZz> also ThreadMessageHandler2
 60 2011-01-31 00:24:21 <jgarzik> get TSH to queue "you received bytes on socket X" messages to TMH, and eliminate TMH's polling loop.
 61 2011-01-31 00:25:16 <ArtForzZz> basically we now have TSH buffering bytes until it has a complete message, putting that in a per-node queue
 62 2011-01-31 00:25:40 <jgarzik> I thought 'have a complete msg' logic was in TMH
 63 2011-01-31 00:25:45 <jgarzik> but maybe I'm mistaken
 64 2011-01-31 00:25:57 <ArtForzZz> you're right
 65 2011-01-31 00:25:58 <ArtForzZz> WTF
 66 2011-01-31 00:26:03 <jgarzik> exactly!
 67 2011-01-31 00:26:38 <ArtForzZz> that makes no sense at all
 68 2011-01-31 00:26:57 <jgarzik> TSH uses select properly, then TMH destroys that with a stupid poll
 69 2011-01-31 00:27:05 <andrew12> gavinandresen: make a libbitcoin!
 70 2011-01-31 00:27:18 <andrew12> and write it in C!
 71 2011-01-31 00:27:19 <andrew12> :P
 72 2011-01-31 00:27:43 <ArtForzZz> either do it single threaded or massively threaded
 73 2011-01-31 00:27:47 <ArtForzZz> but this is just half-assed
 74 2011-01-31 00:27:49 <gavinandresen> C, huh...   kick it old school....
 75 2011-01-31 00:28:06 <andrew12> gavinandresen: it'd be easier to make a plugin/module for other languages if it were in c
 76 2011-01-31 00:28:18 <ArtForzZz> basically either lose TMH or have one TMH per node
 77 2011-01-31 00:29:26 <gavinandresen> andrew12: yeah...  the API could be all C and the guts C++.
 78 2011-01-31 00:29:31 <andrew12> yep
 79 2011-01-31 00:29:35 <jgarzik> well, again, there is nothing _wrong_ with one-TMH, if one chooses the network I/O thread (TSH) + message execution thread (TMH) model.   That is probably the easiest to achieve with current codebase.
 80 2011-01-31 00:29:55 <jgarzik> have TSH figure out message boundaries, then queue "you have a message" msg
 81 2011-01-31 00:30:00 <jgarzik> TMH processes that queue
 82 2011-01-31 00:30:10 <ArtForzZz> yep
 83 2011-01-31 00:30:20 <andrew12> plus refactoring is generally good
 84 2011-01-31 00:30:23 <andrew12> :p
 85 2011-01-31 00:30:42 <ArtForzZz> though I think I'd just lose TMH and roll its functionality into TSH
 86 2011-01-31 00:30:57 <jgarzik> nothing wrong with that, either
 87 2011-01-31 00:31:07 <ArtForzZz> looks simpler from here
 88 2011-01-31 00:31:16 <gavinandresen> refactoring is great if you have a good reason to do it.  If you're just doing it to make the code look prettier, you could probably be doing something more productive.
 89 2011-01-31 00:31:16 <jgarzik> having TSH call ProcessMessage would be fine
 90 2011-01-31 00:31:36 <andrew12> that's true. :p
 91 2011-01-31 00:33:24 <jgarzik> of course, these are pirate rules:  it's not a rule, more like a guideline.
 92 2011-01-31 00:35:25 <ArtForzZz> old-school C coder here
 93 2011-01-31 00:35:43 <ArtForzZz> aka "one function to rule them all"
 94 2011-01-31 00:35:48 <jgarzik> heh
 95 2011-01-31 00:36:25 <Diablo-D3> heh
 96 2011-01-31 00:36:31 <ArtForzZz> calls used to be expensive, and compilers sucked at inlining
 97 2011-01-31 00:36:31 <Diablo-D3> anything more than about 1000 lines is too much
 98 2011-01-31 00:36:51 <Diablo-D3> yeah, now compilers produce code that looks nothing like the code you gave it
 99 2011-01-31 00:36:55 <Diablo-D3> ArtForzZz: and btw
100 2011-01-31 00:37:01 <Diablo-D3> compilers DID listen to inline
101 2011-01-31 00:37:03 <jgarzik> that would eliminate RPC thread
102 2011-01-31 00:37:09 <jgarzik> and TMH
103 2011-01-31 00:37:21 <Diablo-D3> the funny part now is, they dont
104 2011-01-31 00:38:58 <jgarzik> oh well, thoughts for another day.  back to python.
105 2011-01-31 00:44:09 <andrew12> why does bitcoin use a binary protocol?
106 2011-01-31 00:45:27 <andrew12> and why does it do confusing stuff with the endianness?
107 2011-01-31 00:45:40 <jgarzik> oh nice, python can do "have_work = (len(work) > 0)" to assign booleans
108 2011-01-31 00:46:45 <ArtForzZz> yup
109 2011-01-31 00:56:00 <jgarzik> from C and Perl
110 2011-01-31 00:56:04 <echelon> wb joe_1
111 2011-01-31 00:57:13 <Diablo-D3> ArtForzZz: hey
112 2011-01-31 00:57:21 <Diablo-D3> we never discussed the important thing the other day
113 2011-01-31 00:57:28 <Diablo-D3> your kernel massively increases nvidia performance
114 2011-01-31 00:57:35 <ArtForzZz> *shrug*
115 2011-01-31 00:57:43 <ArtForzZz> geuss nvidias compiler is even braindeader than ATIs
116 2011-01-31 00:57:48 <Diablo-D3> its not the hardware choking on m0's kernel
117 2011-01-31 00:57:54 <Diablo-D3> it has to be the compiler
118 2011-01-31 00:57:57 <Diablo-D3> no other way
119 2011-01-31 00:58:44 <Diablo-D3> ArtForzZz: but the funny part is
120 2011-01-31 00:58:53 <Diablo-D3> it drags nvidia significantly forwards
121 2011-01-31 00:58:59 <ArtForzZz> ?
122 2011-01-31 00:59:12 <Diablo-D3> I mean, nvidia was about 4xxx performance
123 2011-01-31 00:59:19 <Diablo-D3> its half way between 4xxx and 5xxx now
124 2011-01-31 00:59:26 <ArtForzZz> errr... no, still is
125 2011-01-31 00:59:55 <ArtForzZz> they're finally getting numbers close to what they *should* be getting
126 2011-01-31 00:59:56 <jgarzik> does python have the equivalent of "perl -cw script.pl", which performs compilation/syntax check but does -not- run the script?
127 2011-01-31 01:00:19 <Diablo-D3> ArtForzZz: oh.
128 2011-01-31 01:00:20 <Diablo-D3> wow
129 2011-01-31 01:00:22 <Diablo-D3> thats shit then
130 2011-01-31 01:00:36 <nanotube> jgarzik: yes, start python, then 'import yourfile'
131 2011-01-31 01:00:39 <nanotube> :)
132 2011-01-31 01:00:40 <ArtForzZz> = with m0s old miner, we saw ~45Mhps on a 260
133 2011-01-31 01:00:55 <jgarzik> nanotube: something on the command line...
134 2011-01-31 01:00:59 <ArtForzZz> while my spreadsheet said ~68Mhps on a 260
135 2011-01-31 01:01:24 <ArtForzZz> and thats for a straightforward double-sha256
136 2011-01-31 01:02:04 <nanotube> jgarzik: well, try http://bytes.com/topic/python/answers/38328-python-switch-syntax-checking
137 2011-01-31 01:03:41 <Diablo-D3> ArtForzZz: I wonder what mine gets on it now
138 2011-01-31 01:04:59 <ArtForzZz> point is, nvidia doesn't have a native rotate, and nothing like ch() either
139 2011-01-31 01:05:29 <ArtForzZz> so minimum ops/hash is pretty much the same between nvidia 2/4/5xx and ati 4xxx
140 2011-01-31 01:14:03 <jgarzik> hmmm.  if I read rpc.cpp correctly, the RPC server does not support persistent connections.
141 2011-01-31 01:39:42 <andrew12> so i wrote the serialization/deserialization functions for ruby-bitcoin.. (well some of them),  but i currently have no idea if they work
142 2011-01-31 01:40:03 <andrew12> but i won't know till tomorrow
143 2011-01-31 01:40:15 <andrew12> if even :P
144 2011-01-31 01:41:50 <joe_1> so it's ready to run on the network?
145 2011-01-31 02:05:32 <andrew12> joe_1: no
146 2011-01-31 02:05:54 <andrew12> lol
147 2011-01-31 02:05:55 <andrew12> not yet
148 2011-01-31 02:06:02 <andrew12> hopefully tomorrow
149 2011-01-31 02:08:17 <joe_1> cool
150 2011-01-31 02:08:54 <joe_1> i'm in the process of trying to understand the bitcoin client code. right now working on all the network protocol junk
151 2011-01-31 02:12:33 <andrew12> yeah
152 2011-01-31 02:12:43 <andrew12> binary protocols are hard
153 2011-01-31 02:12:54 <andrew12> parsing xml/json/whatever is much easier
154 2011-01-31 03:50:36 <nanotube> sipa: thanks for the thought. try out bc,prob :)
155 2011-01-31 03:51:01 <nanotube> ;;bc,prob 1000 2y
156 2011-01-31 03:51:19 <gribble> 0.486818357525
157 2011-01-31 03:56:31 <hacim> nanotube: whats that?
158 2011-01-31 03:56:39 <citizen> yeah, what is that?
159 2011-01-31 03:56:50 <nanotube> gives you probability of finding a block, at X khps, in Y time period
160 2011-01-31 03:57:06 <citizen> ah neato
161 2011-01-31 03:57:16 <citizen> ;;bc,stats
162 2011-01-31 03:57:17 <gribble> Current Blocks: 105428 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 1419 blocks | Next Difficulty In About: 1 week, 2 days, 4 hours, 20 minutes, and 21 seconds | Next Difficulty Estimate: 23657.11415162
163 2011-01-31 03:58:04 <nanotube> ;;bc,poolstats
164 2011-01-31 03:58:05 <gribble> {"hashes_ps": 29151988411, "shares": 46501, "active_workers": 390, "round_duration": "1:54:11", "round_started": "2011-01-31 03:03:54", "shares_cdf": 87.909999999999997, "getwork_ps": 120}
165 2011-01-31 03:58:18 <nanotube> heh, pool's steadily at 30ghps or so.
166 2011-01-31 03:58:27 <citizen> so in a little over a week, it's going to be 7-8% more difficult to find blocks?
167 2011-01-31 04:00:24 <Cusipzzz> citizen: yep
168 2011-01-31 04:02:50 <jgarzik> what's the favorite python method for chopping a string multiple parts, and returning those parts as a list?
169 2011-01-31 04:04:07 <jgarzik> something like str.partition(), but based on length rather than a separator
170 2011-01-31 04:04:53 <nanotube> fixed width split, basically?
171 2011-01-31 04:05:15 <jgarzik> yeah
172 2011-01-31 04:05:41 <nanotube> probably a list comprehension
173 2011-01-31 04:06:19 <nanotube> l = [s[i:i+3] for i in range(0, len(s), 3)]
174 2011-01-31 04:06:30 <nanotube> if you want it by 3's e.g.
175 2011-01-31 04:08:59 <jgarzik> nanotube: neat.  though it will be a while before I actually understand what that's doing :)
176 2011-01-31 04:11:05 <nanotube> it basically takes a for loop... and puts its outputs into a list. :)
177 2011-01-31 04:11:17 <nanotube> more importantly, why do you want to split a string into equal chunks? :)
178 2011-01-31 04:14:37 <jgarzik> nanotube: getting 32-bit chunks of data from getwork
179 2011-01-31 04:14:58 <jgarzik> they're received as as strings, and I'm unpacking them into a list of uint32's
180 2011-01-31 04:15:09 <jgarzik> so that they may be byteswapped
181 2011-01-31 04:15:45 <ArtForzZz> whats wrong with just using struct.unpack on the hex string?
182 2011-01-31 04:16:13 <ArtForzZz> errr... .decode("hex") of the hex string
183 2011-01-31 04:18:46 <ArtForzZz> "6457346535".decode("hex") == 'dW4e5'
184 2011-01-31 04:18:59 <Vladimir> if I were bitcoind developer I would use BERT for all binary (non human readable) serialisation and YAML for text (human readable), and no more troubles with decoding and encoding stuff
185 2011-01-31 04:19:34 <ArtForzZz> can't change the block and tx encoding now
186 2011-01-31 04:19:39 <Vladimir> I know
187 2011-01-31 04:19:46 <Kiba> hmm
188 2011-01-31 04:19:50 <nanotube> save that for bitcoin2 :) heh
189 2011-01-31 04:19:55 <ArtForzZz> yup
190 2011-01-31 04:19:56 <Kiba> I am stuck on one project too long
191 2011-01-31 04:20:21 <Kiba> I need to practice drawing all kind of things
192 2011-01-31 04:21:48 <Vladimir> thank's god, though, that there is no XML anywhere in sight, lol
193 2011-01-31 04:23:24 <mrb_> ;;rate Vladimir 1
194 2011-01-31 04:23:25 <gribble> Error: For identification purposes, you must have a freenode cloak to use the rating system.
195 2011-01-31 04:25:38 <mrb_> ;;rate Keefe 0
196 2011-01-31 04:25:42 <gribble> Error: Rating must be in the interval [-10, 10] and cannot be zero.
197 2011-01-31 04:26:38 <Keefe> :o
198 2011-01-31 04:27:09 <mrb_> ;;rate Keefe 1
199 2011-01-31 04:27:42 <gribble> Rating entry successful. Use the 'getrating' command to view Keefe's new rating.
200 2011-01-31 04:28:18 <mrb_> ;;unrate Keefe
201 2011-01-31 04:28:51 <gribble> Successfully removed your rating for Keefe.
202 2011-01-31 04:29:09 <mrb_> just trying to see if rating works...
203 2011-01-31 04:29:36 <Keefe> ah :)
204 2011-01-31 04:29:40 <mrb_> I guess I should use gribble for that
205 2011-01-31 04:29:51 <mrb_> ;;rate gribble 42
206 2011-01-31 04:29:56 <gribble> Error: Rating must be in the interval [-10, 10] and cannot be zero.
207 2011-01-31 04:30:35 <mrb_> ;;rate gribble 10
208 2011-01-31 04:30:39 <gribble> Rating entry successful. Use the 'getrating' command to view gribble's new rating.
209 2011-01-31 04:30:42 <mrb_> ;;unrate gribble
210 2011-01-31 04:30:48 <gribble> Successfully removed your rating for gribble.
211 2011-01-31 04:30:59 <nanotube> mrb_: you could also use #bitcoin-otc for that. :)
212 2011-01-31 04:31:02 <nanotube> or pm with gribble. dunno how much the -dev folks enjoy the rating spam. ;)
213 2011-01-31 04:31:06 <mrb_> oh right I could pm gribble
214 2011-01-31 04:44:27 <hacim> slush: I'm getting a lot of 'invalid or stale'
215 2011-01-31 04:45:33 <hacim> well, 3 in the last 44 minutes, but at 20+ yesterday
216 2011-01-31 04:48:56 <dooglus> jgarzik: this works, too: >>> re.findall("...", "hello world!")
217 2011-01-31 05:16:50 <echelon> ducki2p, what's the matter.. no i2p gateway for freenode? :P
218 2011-01-31 05:17:17 <ducki2p> ;)
219 2011-01-31 05:17:41 <nanotube> echelon: actually there is, on #bitcoin-discussion :)
220 2011-01-31 05:19:21 <echelon> hmm?
221 2011-01-31 05:20:09 <echelon> a relay bot of some sort?
222 2011-01-31 05:20:16 <nanotube> yes :)
223 2011-01-31 05:20:24 <echelon> ah k
224 2011-01-31 06:33:36 <slush> hacim: read the forum
225 2011-01-31 06:33:44 <slush> hacim: then open #bitcoin-monitor
226 2011-01-31 06:34:04 <slush> hacim: and when your 'invalid or stale' fit with times of new blocks, then it is OK
227 2011-01-31 07:16:57 <jgarzik> l = [0,1,2,4,8,16,32,64]
228 2011-01-31 07:16:58 <jgarzik> s = struct.pack('@IIIIIIII', l)
229 2011-01-31 07:17:05 <jgarzik> struct.error: pack requires exactly 8 arguments
230 2011-01-31 07:23:24 <strattog> jgarzik: struct.pack('@IIIIIIII', *l)
231 2011-01-31 08:57:39 <gribble> (bc,prob <an alias, at least 1 argument>) -- Alias for "math calc 1-exp(-$1*1000 * [seconds $*] / (2**32* [bc,diff]))".
232 2011-01-31 08:57:39 <sipa> ;;bc,prob
233 2011-01-31 08:58:23 <sipa> ;;bc,prob 1230000 8h
234 2011-01-31 08:58:24 <gribble> 0.312495482373
235 2011-01-31 08:58:56 <gribble> 0.675042422228
236 2011-01-31 08:58:56 <sipa> ;;bc,prob 1230000 1d
237 2011-01-31 10:10:50 <BostX> hi ppl
238 2011-01-31 10:12:14 <BostX> on http://bitcoinwatch.com/ I see a steady increase of USD invested to bitcoins
239 2011-01-31 10:12:20 <BostX> how do u explain that
240 2011-01-31 10:12:21 <BostX> ?
241 2011-01-31 10:41:15 <satamusic> greed.
242 2011-01-31 10:42:03 <Diablo-D3> Greedo-san.
243 2011-01-31 10:42:12 <Diablo-D3> hrm, or Greeed
244 2011-01-31 10:42:14 <Diablo-D3> that would be redundantly delicious
245 2011-01-31 11:49:58 <noagendamarket> I blame the mob BostX
246 2011-01-31 11:53:58 <BostX> noagendamarket: hmm... never thought the mob is that rich
247 2011-01-31 11:54:07 <opus_> yo
248 2011-01-31 11:54:32 <opus_> bjam is a bitch to cross compile
249 2011-01-31 11:56:20 <noagendamarket> The russian mob owns hollywood
250 2011-01-31 12:03:46 <UukGoblin> ha, just read about BitX
251 2011-01-31 12:03:55 <UukGoblin> I thought of exactly the same thing on my own
252 2011-01-31 12:04:22 <UukGoblin> the bitcoin's block chain could easily be used for a multitude of other uses, e.g. signing and timestamping MMORPG events
253 2011-01-31 12:05:53 <BostX> UukGoblin: post a link pls
254 2011-01-31 12:09:25 <UukGoblin> http://www.bitcoin.org/smf/index.php?topic=1790.0
255 2011-01-31 12:17:08 <opus_> ?
256 2011-01-31 12:17:14 <opus_> whats bitx
257 2011-01-31 12:18:03 <opus_> oh linkie
258 2011-01-31 12:19:28 <opus_> What happeneds when they block IP address.. ?
259 2011-01-31 12:20:02 <opus_> "You can only go to facebook.com" - Nancy Peloski 2012
260 2011-01-31 12:25:28 <citizen> lol
261 2011-01-31 12:26:01 <citizen> nancy pelosi's facebook is probably worse than tubgirl/lemonparty
262 2011-01-31 12:27:14 <Diablo-D3> citizen: her real one? "yes", but shes too old to actually even know what computers are
263 2011-01-31 12:27:47 <Diablo-D3> her fake one is just ran by her PR team
264 2011-01-31 12:41:03 <MT`AwAy> :o
265 2011-01-31 12:47:41 <grondilu> what's up with biddingpond ?
266 2011-01-31 12:51:06 <hacim> two blocks within 38 seconds of each other
267 2011-01-31 12:51:25 <hacim> slush: ok, that makes sense, thanks for pointing that out.
268 2011-01-31 12:52:27 <hacim> slush: round times are really high aren't they? in the last 6hrs, even with 700 some shared, i've only received a very small amount of reward :o
269 2011-01-31 13:02:49 <citizen> is slush around?
270 2011-01-31 13:20:14 <gavinandresen> Anybody here want to help generate blocks on the new testnet chain?  (means grabbing a patch, recompiling, running and reporting any weirdnesses caused by your old testnet wallet/block chain suddenly becoming invalid)
271 2011-01-31 13:22:47 <tcatm> gavinandresen: Where's the patch? :)
272 2011-01-31 13:23:13 <gavinandresen> I haven't pushed it yet, will do that in a couple minutes
273 2011-01-31 13:26:25 <gavinandresen> tcatm: https://github.com/bitcoin/bitcoin/pull/53
274 2011-01-31 13:36:35 <gavinandresen> tcatm:  new testnet height should be 62... let me double-check and make sure my firewall is open to testnet connections....
275 2011-01-31 13:38:05 <tcatm> It still shows the old height (28385). What's your testnet node's IP?
276 2011-01-31 13:38:51 <gavinandresen> 96.240.197.31
277 2011-01-31 13:39:38 <gavinandresen> It should show height=0 and a whole bunch of orphan blocks if you have the new genesis block
278 2011-01-31 13:42:28 <tcatm> strange. should it detect the changed genesis block during startup?
279 2011-01-31 13:43:21 <gavinandresen> Yeah, but I there is an optimization stopping it:  if (pindex->nHeight < nBestHeight-2500 && !mapArgs.count("-checkblocks"))
280 2011-01-31 13:43:54 <gavinandresen> ... it only checks the last 2500 blocks to see if they're valid.
281 2011-01-31 13:44:35 <tcatm> still 28385
282 2011-01-31 13:46:04 <gavinandresen> Right, so should we add code to check for a changed genesis block (if testnet), or just tell people to remove blk*.dat.....
283 2011-01-31 13:46:26 <tcatm> How hard would it be to check the hash of the genesis block on every start?
284 2011-01-31 13:46:37 <gavinandresen> Not hard
285 2011-01-31 13:46:52 <gavinandresen> (famous last words)
286 2011-01-31 13:49:47 <UukGoblin> hmm I wonder if compiling with -fprofile-generate and -fprofile-use would speed up the stock's client generation :-]
287 2011-01-31 13:50:54 <tcatm> UukGoblin: unlikely
288 2011-01-31 13:51:30 <UukGoblin> yeah I wouldn't expect miracles with it
289 2011-01-31 13:52:22 <tcatm> At least my 4way miner is optimized to only use the cache
290 2011-01-31 13:53:37 <citizen> are you talking about CPU mining?
291 2011-01-31 13:54:06 <tcatm> yes
292 2011-01-31 13:54:43 <Kiba> hey bitcoiners
293 2011-01-31 13:55:38 <tcatm> hey Kiba
294 2011-01-31 14:03:33 <gribble> 0.51169600304
295 2011-01-31 14:03:33 <sipa> ;;bc,prob 1255000 15h
296 2011-01-31 14:04:05 <sipa> ;;bc,prob 1257000 15h
297 2011-01-31 14:04:06 <gribble> 0.512253492791
298 2011-01-31 14:05:50 <UukGoblin> hm that's cool
299 2011-01-31 14:06:02 <UukGoblin> ;;bc,prob 2000000 24h
300 2011-01-31 14:06:03 <gribble> 0.839223627479
301 2011-01-31 14:06:10 <gribble> 0.99999998846
302 2011-01-31 14:06:10 <UukGoblin> ;;bc,prob 2000000 240h
303 2011-01-31 14:12:04 <necrodearia> ooh, bitcoin.org/smf looks different
304 2011-01-31 14:13:10 <tcatm> [6~[6~[6~ga
305 2011-01-31 14:15:24 <tcatm> gavinandresen: listaccounts outputs negative balance for most accounts
306 2011-01-31 14:16:08 <gavinandresen> tcatm:  Oooh.... yeah....  hmmm.
307 2011-01-31 14:17:46 <gavinandresen> That's because you spent a whole bunch of now-invalid coins.
308 2011-01-31 14:18:03 <tcatm> yep
309 2011-01-31 14:21:44 <gavinandresen> Yeah, I think recommending that you remove your testnet data directory (nuking your wallet and the old blk files) is the right thing to do.
310 2011-01-31 14:29:04 <luke-jr> hey, my bitcoins actually sold
311 2011-01-31 14:29:08 <luke-jr> 0.4799 USD ea
312 2011-01-31 14:32:12 <tcatm> gavinandresen: btw, do we really need to create "Your Address" account? I thought "" was the default now
313 2011-01-31 14:33:19 <gavinandresen> I don't care enough to change that code.
314 2011-01-31 14:47:57 <davout> gavinandresen: has the default account changed name again ?
315 2011-01-31 14:48:10 <gavinandresen> davout: no
316 2011-01-31 14:48:34 <gavinandresen> The "" account is the default for generated coins.
317 2011-01-31 14:48:58 <gavinandresen> The "Your Address" is the GUI's default for receiving coins.
318 2011-01-31 14:49:58 <luke-jr> gavinandresen: it seems people keep suggesting the same solutions to the sub-cent issue :p
319 2011-01-31 14:50:18 <luke-jr> my RPC v1 = sipa's Idea 2 = vladimir's "why not"
320 2011-01-31 14:50:29 <luke-jr> my RPC v2 = your counter-proposal
321 2011-01-31 14:50:31 <luke-jr> err
322 2011-01-31 14:50:37 <luke-jr> my RPC v0 w/o rounding*
323 2011-01-31 14:50:59 <gavinandresen> Great minds think alike?
324 2011-01-31 14:51:15 <luke-jr> :
325 2011-01-31 14:51:18 <gavinandresen> Either that or herd mentality
326 2011-01-31 14:51:41 <luke-jr> perhaps, I was just commenting on how each time these are proposed, it seems like an argument
327 2011-01-31 14:51:49 <luke-jr> even though the ideas are basically the same thing :p
328 2011-01-31 14:54:31 <gavinandresen> basically.... but I don't think supporting multiple versions of the RPC interface is a good idea, just because it gives some people the warm fuzzies to send 100000000. instead of 1.0
329 2011-01-31 14:55:05 <gavinandresen> (or send 1 instead of 1e-8)
330 2011-01-31 14:55:13 <sipa> gavinandresen: the command-line interface and GUI could do transformation from/to decimal-BTC notation
331 2011-01-31 14:55:36 <gavinandresen> sipa:  uh-huh.  And JSON still defines numbers to be double-precision floats
332 2011-01-31 14:55:36 <sipa> on the other hand, your proposal seems a combination... solve the issue of sub-cent precision and make the JSON interface support all transactions, while still keeping it human readable and (probably) not breaking backward-compatibility
333 2011-01-31 14:56:57 <sipa> but i still don't like relying on covertion from and to floating point to transfer something which should be represented as integers internally
334 2011-01-31 14:57:03 <sipa> *conversion
335 2011-01-31 14:57:16 <gavinandresen> Human readable protocols are extremely handy...  I think its a big reason HTTP is a huge success.
336 2011-01-31 14:57:45 <sipa> yes, that's a reasonable argument
337 2011-01-31 14:57:59 <sipa> as is not wanting to break backward compatibility
338 2011-01-31 14:58:02 <gavinandresen> sipa:  so don't convert to a float, just take the string and move the decimal point 8 places to the right.
339 2011-01-31 14:58:38 <sipa> but using any json library will parse it as a float for you - expect when encoding it as a string
340 2011-01-31 14:58:45 <luke-jr> gavinandresen: people shouldn't be using the RPC interface anyway
341 2011-01-31 14:59:10 <luke-jr> gavinandresen: and it's almost inevitable that something will require a revision bump at some point
342 2011-01-31 14:59:28 <sipa> his proposal doesn't
343 2011-01-31 14:59:35 <luke-jr> JSON doesn't define numbers ot be double-precision floats.
344 2011-01-31 15:00:52 <gavinandresen> http://www.ietf.org/rfc/rfc4627.txt
345 2011-01-31 15:01:02 <sipa> true - but it defines exponential representation, and allows implementations to limit its supported range
346 2011-01-31 15:01:04 <gavinandresen> Normative reference:  ECMA, aka JavaScript....
347 2011-01-31 15:01:13 <luke-jr> hmm, it wasn't on the site, oh well
348 2011-01-31 15:01:30 <luke-jr> still, floating-point can represent all of the valid bitcoin integer values losslessly ;)
349 2011-01-31 15:01:42 <luke-jr> and it *can't* do the same with fractional BTC values
350 2011-01-31 15:01:48 <luke-jr> (short of rounding)
351 2011-01-31 15:02:14 <gavinandresen> why yes, yes it can.  It is almost as if Satoshi chose 2.1 quadrillion bittie-coins on purpose......
352 2011-01-31 15:02:19 <luke-jr> also, your linked spec doesn't specify floating-point either
353 2011-01-31 15:02:45 <sipa> gavinandresen: it think luke-jr means exact representation, not exact recovery after rounding
354 2011-01-31 15:03:04 <luke-jr> sipa: right
355 2011-01-31 15:03:15 <sipa> it's obvious that a double suffices for every possible bitcoin amount
356 2011-01-31 15:03:19 <sipa> in terms of information content
357 2011-01-31 15:03:54 <luke-jr> with floating-point BTC representation, you have to consider an error factor when doing math and printing (into a JSON string, even)
358 2011-01-31 15:03:59 <gavinandresen> why are we spending so much time talking about all this, when we could be talking about whether emacs or vi is better?
359 2011-01-31 15:04:25 <luke-jr> dunno, I just figured using base units was the obvious solution :p
360 2011-01-31 15:04:45 <luke-jr> didn't expect an argument over it
361 2011-01-31 15:04:58 <gavinandresen> obvious, but not best in my humble opinion (given constraints of backwards compatibility, etc)
362 2011-01-31 15:05:11 <sipa> i agree with luke-jr here, but i'm not sure it's worth sacrificing backward compatibilty
363 2011-01-31 15:05:18 <luke-jr> gavinandresen: as noted, my implementation of it is 100% backward-compatible
364 2011-01-31 15:05:31 <gavinandresen> ... except for the GUI which you don't care about.....
365 2011-01-31 15:05:46 <luke-jr> afaik, the GUI is unaffected by the changes
366 2011-01-31 15:05:51 <luke-jr> it only affects RPC
367 2011-01-31 15:06:22 <luke-jr> the GUI's display precision bug is unrelated to RPC interaction IMO
368 2011-01-31 15:06:33 <gavinandresen> No, one of your compatible changes added sub-cent throwaway change to the fee reported.  Which would affect the GUI
369 2011-01-31 15:06:47 <luke-jr> gavinandresen: that's moved into another branch now, and a different topic
370 2011-01-31 15:06:55 <gavinandresen> (GUI users could get a very confusing "this transaction needs a fee of 0.00 bitcoins)
371 2011-01-31 15:07:15 <luke-jr> gavinandresen: I agree, before hitting a stable release, that particular branch needs some GUI follow-up
372 2011-01-31 15:07:49 <gavinandresen> Cool.  I'd like to see a set of sub-cent-precision patches next release.
373 2011-01-31 15:08:12 <luke-jr> but the avoid-subcent-fees change ('master' branch), and base unit RPC ('neutral' branch) seem like they should be easy clean merges to me
374 2011-01-31 15:08:34 <gavinandresen> Can anybody here easily spin up a Windows development (mingw) virtual environment?
375 2011-01-31 15:08:46 <luke-jr> I can, but not with wx
376 2011-01-31 15:08:52 <luke-jr> I don't even have/want wx on my native install :p
377 2011-01-31 15:09:07 <necrodearia> <gavinandresen> why are we spending so much time talking about all this, when we could be talking about whether emacs or vi is better? nano ftw
378 2011-01-31 15:09:53 <gavinandresen> Not deleted, replaced.
379 2011-01-31 15:09:54 <luke-jr> gavinandresen: git pull git://gitorious.org/bitcoin/bitcoin.git master # should get ONLY the avoid-subcent-fee fix
380 2011-01-31 15:10:49 <gavinandresen> My priority right now is getting a .20 release out; no pulls of anything new at this point, just build fixes or critical bug fixes.
381 2011-01-31 15:11:05 <luke-jr> IMO, the avoid-subcent-fee fix is a critical fix, but oh well
382 2011-01-31 15:11:30 <luke-jr> since it wastes bitcoins behind your back
383 2011-01-31 15:12:07 <tcatm> even if it's critical we have lots of time to fix bugs until something like 1.0.0
384 2011-01-31 15:12:29 <luke-jr> what features are considered 1.0.0?
385 2011-01-31 15:12:37 <luke-jr> I don't see why it would be "time" based
386 2011-01-31 15:17:45 <gavinandresen> My list of features for a 1.0 release would include:   encrypt/password-protect wallet.  Import/export wallet.   Lightweight (header-only) client, maybe with block-headers shipped as part of install.  Much more polished GUI.
387 2011-01-31 15:18:35 <gavinandresen> Oh, and a bitcoin.org redesign/rewrite, to target non-geeks.
388 2011-01-31 15:19:00 <luke-jr> encrypt wallet is definitely a must-have :D
389 2011-01-31 15:19:29 <luke-jr> and store the decrypted data in protected RAM
390 2011-01-31 15:19:32 <luke-jr> (eg, non-swappable)
391 2011-01-31 15:20:25 <gavinandresen> It shouldn't be stored at all; should be read from disk, decrypted, used to sign transaction, then overwritten and memory freed.
392 2011-01-31 15:20:26 <luke-jr> gavinandresen: btw, if autotools isn't getting in for .20, do you want me to git-ify qmake support?
393 2011-01-31 15:20:38 <gavinandresen> talk to jgarzik
394 2011-01-31 15:20:46 <luke-jr> gavinandresen: you're the one doing merging :P
395 2011-01-31 15:20:55 <luke-jr> /releases
396 2011-01-31 15:20:59 <gavinandresen> I delegated linux build stuff to him
397 2011-01-31 15:21:13 <luke-jr> ok, it just sounded like you were on the verge of doing a release today
398 2011-01-31 15:21:18 <gavinandresen> I am
399 2011-01-31 15:21:20 <luke-jr> and that stuff is obviously not there yet
400 2011-01-31 15:21:43 <gavinandresen> (well, doing a release candidate)
401 2011-01-31 15:21:58 <luke-jr> so I was just thinking throw in qmake support as a quick temporary thing, to be replaced by automake next release
402 2011-01-31 15:22:07 <gavinandresen> Yeah... no.
403 2011-01-31 15:22:22 <luke-jr> k
404 2011-01-31 15:22:40 <luke-jr> (it wouldn't affect any of the existing files at all, ofc& but your call)
405 2011-01-31 15:22:43 <sipa> now that i think about it... are there still hard-coded paths in the makefile?
406 2011-01-31 15:23:23 <sipa> i did, by replacing the hard-coded paths to calls to wx-config
407 2011-01-31 15:23:48 <tcatm> sipa: there's a pull request to fix them
408 2011-01-31 15:23:50 <luke-jr> well, I don't have wx-config ;)
409 2011-01-31 15:23:51 <gavinandresen> Makefile tweaks aren't critical bugs that need fixing, unless they're preventing a release candidate from getting built.  ANd they're not.
410 2011-01-31 15:24:14 <sipa> ok
411 2011-01-31 15:24:17 <luke-jr> gavinandresen: you did say just *build fixes or* critical bug fixes :P
412 2011-01-31 15:24:38 <sipa> i don't care about whether those things are fixed in the coming release or a later one
413 2011-01-31 15:24:42 <gavinandresen> Yeah, "build fixes" means stuff that will make the build work.  Not stuff that makes the build more pleasant.
414 2011-01-31 15:24:46 <luke-jr> ah
415 2011-01-31 15:25:01 <gavinandresen> (I'm all for making the build more pleasant, just not today)
416 2011-01-31 15:25:34 <luke-jr> well, I still vote to merge avoid-subcent-precision for .20, anyhow. do or don't, I'm still using git myself. :p
417 2011-01-31 15:26:13 <tcatm> what makes .20 so special you can't wait for .21 or .22?
418 2011-01-31 15:26:24 <CIA-98> bitcoin: Gavin Andresen master * rbd7d914 / main.cpp : new checkpoint at block 105,000 - http://bit.ly/g6zmyV
419 2011-01-31 15:26:40 <nanotube> woo, new checkpoint
420 2011-01-31 15:27:28 <gavinandresen> 105K was a very modest block, for being so historic... (just the one coinbase transaction in it)
421 2011-01-31 15:27:51 <luke-jr> tcatm: I run git, so it isn't for me. It just seems wrong to throw away others' moneys without even a warning
422 2011-01-31 15:28:16 <luke-jr> especially when it's completely unnecessary
423 2011-01-31 15:28:26 <tcatm> luke-jr: it still says "beta" in the code so it's okay for now
424 2011-01-31 15:28:35 <gavinandresen> luke-jr: if it affected more than .1% of bitcoin users, or if it cost them more than a fraction of a penny then it would be serious.
425 2011-01-31 15:28:37 <luke-jr> true I guess
426 2011-01-31 15:28:55 <luke-jr> gavinandresen: long-term, it could cost a lot
427 2011-01-31 15:29:24 <tcatm> Though if it's really a problem for anyone, contact me and I'll send them 0.01 BTC for their subcent loss :P
428 2011-01-31 15:30:07 <luke-jr> tcatm: will you send them 0.01 BTC by today's value, to them in 2 years when it's worth 10 BTC? :P
429 2011-01-31 15:30:20 <sipa> i'm just glad there is a solution
430 2011-01-31 15:30:30 <sipa> whether or not it's incorporated today
431 2011-01-31 15:30:36 <tcatm> luke-jr: yes
432 2011-01-31 15:31:08 <luke-jr> by that I mean, 0.01 BTC today will be worth $100 in the future if they had saved it.
433 2011-01-31 15:31:33 <gavinandresen> If we go up in value a factor of 1,000, minimum transaction size will have to be changed, and tcatm can send them exactly how many they lost.
434 2011-01-31 15:33:06 <gavinandresen> Right.... so I'm not hearing that anybody can easily spin up a bitcoin-on-windows virtual build environment.   I'm going to eat lunch, then spin up an EC2 instance to get that going....
435 2011-01-31 15:36:07 <luke-jr> gavinandresen: the problem isn't the Windows build environment, it's wx :p
436 2011-01-31 15:36:14 <luke-jr> if we had a Qt GUI, it'd be trivial
437 2011-01-31 15:36:42 <gavinandresen> Do you know Qt programming?  Go for it!
438 2011-01-31 15:36:49 <gavinandresen> (shouldn't be too hard, right?)
439 2011-01-31 15:37:03 <nanotube> luke-jr: MT`AwAy is working on a qbitcoin client
440 2011-01-31 15:37:07 <nanotube> might care to join his effort
441 2011-01-31 15:37:26 <luke-jr> nanotube: I would, but he won't publish code yet 9
442 2011-01-31 15:37:43 <nanotube> well, talk to him, maybe he'd like some help...
443 2011-01-31 15:37:49 <luke-jr> gavinandresen: I wouldn't know what to implement, having never seen the wx client
444 2011-01-31 15:37:50 <tcatm> luke-jr: there's a nogui branch on my github that removes any wx code. You could easily add a Qt GUI to it
445 2011-01-31 15:38:09 <luke-jr> IIRC he's waiting until it's more ready or something
446 2011-01-31 15:38:28 <nanotube> mmm
447 2011-01-31 15:38:54 <nanotube> tcatm: mm isn't that just bitcoind? :)
448 2011-01-31 15:39:34 <tcatm> compiled, yes. the code is a little bit cleaner, though
449 2011-01-31 15:40:10 <nanotube> nice. maybe it'll become the new bitcoind heh.
450 2011-01-31 15:49:22 <tcatm> Mhm. Maybe it would be a good idea to replicate the wxGUI a standalone application connecting via RPC?
451 2011-01-31 15:52:46 <sipa> it can catch SIGPIPE and reconnect, but i see no reason why a pipe is preferable to a socket
452 2011-01-31 16:01:09 <molecular> ^ if anyone is interested I can clean up and make a diff
453 2011-01-31 16:20:05 <midnightmagic> it's not currently possible with existing code to replicate the GUI via the rpc interface.
454 2011-01-31 16:21:08 <midnightmagic> but it would be very cool if it could be done..!
455 2011-01-31 16:21:32 <gavinandresen> midnightmagic: what's missing?
456 2011-01-31 16:25:05 <luke-jr> gavinandresen: probably a way to warn about fees
457 2011-01-31 16:25:22 <luke-jr> at least
458 2011-01-31 16:25:34 <luke-jr> IMO, the current send API is probably too high-level in general really
459 2011-01-31 16:25:59 <luke-jr> there should be one call to calculate necessary fees, and another to actually send a tx with fee explicitly specified
460 2011-01-31 16:26:22 <gavinandresen> Yeah.... that's non-trivial, because you'd have to lock the coins involved between RPC calls.
461 2011-01-31 16:26:37 <luke-jr> but QBitCoin project is already inventing new RPC calls for these issues, so probably shouldn't reinvent MT`AwAy's work
462 2011-01-31 16:27:01 <luke-jr> gavinandresen: no? fees (usually) don't care about what coins are used
463 2011-01-31 16:27:14 <luke-jr> the only exception is throwaway change
464 2011-01-31 16:27:25 <gavinandresen> Sure they do-- fees depend on transaction size, and tx size depends on which/how many inputs there are
465 2011-01-31 16:27:32 <luke-jr> ah, true
466 2011-01-31 16:28:09 <luke-jr> in fact, it would be an improvement if the "avoid subcent throwaway" code took that into consideration
467 2011-01-31 16:28:17 <luke-jr> but that's probably meaning a complete rewrite of stuff
468 2011-01-31 16:33:10 <midnightmagic> gavinandresen: things like the -printblocktree (which I guess isn't in the GUI anyway) ways of accessing the addressbook, listing not-yet-confirmed blocks, etc, last I checked.. unless you've managed to add all that in trunk.
469 2011-01-31 16:33:58 <midnightmagic> which would be very exciting. :)
470 2011-01-31 16:34:47 <gavinandresen> midnightmagic:  File issues at github so we don't forget what is missing-- not-yet-confirmed blocks is a good one (-printblocktree isn't-- that's undocumented, right?)
471 2011-01-31 16:36:05 <midnightmagic> yes, definitely undocumented. just a way of printing the entire block tree, ands then beside each block whether it belongs to your wallet or not (prints a "mine") next to it.
472 2011-01-31 16:36:24 <midnightmagic> is the github the authoritative? who has keys to the subversion repo?
473 2011-01-31 16:37:05 <gavinandresen> github is the integration tree.  svn repo will be the 'stable' repo; Satoshi and I have commit access to the subversion repo
474 2011-01-31 16:37:44 <midnightmagic> very good then. have we heard publically from satoshi since he "disappeared" yet?
475 2011-01-31 16:38:09 <midnightmagic> (to impose on your time a tad longer..! :-)
476 2011-01-31 16:39:55 <gavinandresen> "publically" ?  I've had email conversations with him in the last couple weeks-- he says he's busy.
477 2011-01-31 16:40:35 <midnightmagic> sometimes i get the impression he's testing to see how well the network survives without him
478 2011-01-31 16:40:53 <gavinandresen> Could be.  I think we're doing OK
479 2011-01-31 16:41:03 <midnightmagic> i agree.
480 2011-01-31 16:42:09 <midnightmagic> on the other hand, we haven't had a concrete example/prosecution of money laundering through bitcoin yet and buddy over in #bitcoin-mining (I think) is talking about deliberately attracting international money launderers. I'm pretty sure that's good and beyond just bringing in wikileaks. :)
481 2011-01-31 16:42:49 <midnightmagic> anyway, thanks for pointing me towards github, i appreciate it!
482 2011-01-31 16:43:26 <sipa> wow, mtgox almost reached $0.5
483 2011-01-31 16:44:03 <midnightmagic> according to the front page, it did.
484 2011-01-31 16:44:14 <sipa> 0.4997
485 2011-01-31 16:44:19 <sipa> not 0.50 :)
486 2011-01-31 16:44:31 <midnightmagic> chart says 0.5000, what are you looking at?
487 2011-01-31 16:44:57 <sipa> #bitcoin-market, but the chart rounds, i think
488 2011-01-31 16:45:13 <midnightmagic> there's another dot right under the top one that does say 0.4997
489 2011-01-31 16:45:22 <sipa> hmm
490 2011-01-31 16:45:46 <sipa> there's (still) an ask left at 0.4999
491 2011-01-31 16:45:58 <sipa> but it could be very recent of course
492 2011-01-31 16:46:08 <midnightmagic> hrm.
493 2011-01-31 16:47:35 <sipa> hmm, there's actually been a 0.5 trade, according to http://mtgox.com/code/data/getTrades.php
494 2011-01-31 16:48:53 <midnightmagic> http://mtgox.com/code/data/getTrades.php
495 2011-01-31 16:49:05 <midnightmagic> ah you beat me to it.
496 2011-01-31 16:52:53 <sipa> heh
497 2011-01-31 16:52:55 <sipa> it's gone again
498 2011-01-31 16:53:10 <sipa> oh, nvm
499 2011-01-31 16:54:22 <midnightmagic> :)
500 2011-01-31 17:00:35 <grondilu> woaw we're back at 50cts, guys !
501 2011-01-31 17:06:05 <grondilu> http://www.youtube.com/watch?v=5qm8PH4xAss  :)
502 2011-01-31 17:45:58 <gribble> 22012.4941572
503 2011-01-31 17:45:58 <sipa> ;;bc,diff
504 2011-01-31 17:47:40 <bitanarchy> I bet the miners are certelizing to pull the price up.... which is actually good, because it will motivate more people to start gpu mining...
505 2011-01-31 17:55:32 <bitanarchy> Or lots of new investors are moving in... which also pulls up the price...
506 2011-01-31 18:00:51 <slush> bitanarchy: buying bitcoins is still safest way how to get in
507 2011-01-31 18:01:10 <slush> bitanarchy: mining is more risky and need time & knowledge
508 2011-01-31 18:02:06 <bitanarchy> slush: If that is the case then that means that the cartel will stay..... which is a reason not to invest in bitcoin...
509 2011-01-31 18:02:49 <slush> why? Anybody can set up mining rig. But buying is easier for my 60 years old uncle
510 2011-01-31 18:04:07 <slush> he will never study how to set up ati card. It is the same as I won't study how to find a gold. I'm just buying it at store
511 2011-01-31 18:04:35 <newsham> i wonder why the growth in number of miners isnt driving the price of btc down
512 2011-01-31 18:04:51 <newsham> arent they selling the coins they mine?
513 2011-01-31 18:05:13 <slush> well, I'm selling it almost directly
514 2011-01-31 18:05:57 <slush> forum and irc is full of miners, but I doubt it is majority of bitcoiners
515 2011-01-31 18:06:04 <bitanarchy> newsham: more miners means less cartels... which means a fair price
516 2011-01-31 18:06:22 <newsham> can you explain to me what a fair price for btc is?
517 2011-01-31 18:06:37 <newsham> based on some fundamentals analysis?
518 2011-01-31 18:06:42 <bitanarchy> newsham: the price you would get without cartels :-)
519 2011-01-31 18:07:26 <newsham> also are you sure there are more miners and not just more powerful cartels?
520 2011-01-31 18:07:37 <newsham> difficulty doesnt say how many of em there are
521 2011-01-31 18:08:53 <bitanarchy> If someone invents an efficient way of mining he pulls up difficulty and get to generate each block...
522 2011-01-31 18:09:38 <TD> why would more miners drive down the price of bitcoins?
523 2011-01-31 18:09:44 <TD> number of coins minted has no relation to mining capacity
524 2011-01-31 18:09:49 <TD> (over the long run)
525 2011-01-31 18:09:52 <newsham> since BTC doesnt trade for much right now besides dollars and currency, the only fundamentals I canthink of are comparing BTC to energy costs.
526 2011-01-31 18:10:28 <newsham> TD: if people were mining BTC to sell to get dollars, then it would be selling pressure.
527 2011-01-31 18:10:38 <newsham> but you're right, the selling volume from mining would be fairly constant I guess.
528 2011-01-31 18:10:52 <TD> right. i think most miners were already selling
529 2011-01-31 18:11:09 <TD> it's also possible the value of bitcoins on the exchange is a speculative bubble
530 2011-01-31 18:11:20 <newsham> but at any rate, you have miners who are constantly selling some fraction of 50BTC * 6/hr * 24hrs per day.
531 2011-01-31 18:11:35 <newsham> who is constantly on the buy side to overwhelm the sells and drive up the price?
532 2011-01-31 18:11:46 <newsham> yah, seems like a speculative bubble to me
533 2011-01-31 18:12:08 <newsham> given that I dont see much other use for buying BTC at the moment (hopefully that will change)
534 2011-01-31 18:12:16 <bitanarchy> Maybe the miners got enough of the hard work, got together and came up with a plan...
535 2011-01-31 18:14:54 <bitanarchy> most people are sitting on their coins anyway... so the miners are the main sellers, besides the speculators who are busy with arbitrage, etc..
536 2011-01-31 18:16:04 <newsham> what are they arbitraging?
537 2011-01-31 18:16:33 <newsham> also how do you know that most people are sitting on their coins?
538 2011-01-31 18:16:36 <bitanarchy> between bitcoin-central and mtgox, and bitcoin-market maybe...
539 2011-01-31 18:16:40 <slush> it's too bad that bitcoin forum is from 90% about mining. But it is pretty hard to build business on top of Bitcoins when there's such small market
540 2011-01-31 18:18:07 <slush> it still such hard to get bitcoins for common people. Or is there any more complete list than bitcoinbuy.com ?
541 2011-01-31 18:18:22 <slush> There is only few people, it would be great to have at least one exchanger in every country
542 2011-01-31 18:18:45 <newsham> *nod* took several weeks for me to get $ to BTC
543 2011-01-31 18:19:07 <newsham> also there's no great incentive to buy BTC at the moment.  there's not a lot you can do with BTC that you cant do directly with $ right now.
544 2011-01-31 18:19:15 <bitanarchy> newsham: why do long?
545 2011-01-31 18:19:26 <newsham> bit; *shrug*
546 2011-01-31 18:19:52 <citizen> bitcoins are more collectible than dollars :D
547 2011-01-31 18:19:54 <bitanarchy> newsham: if it is not new investors, what is driving up the price of bitcoin?
548 2011-01-31 18:19:56 <citizen> there are such limited dollars
549 2011-01-31 18:19:58 <citizen> errr
550 2011-01-31 18:20:00 <citizen> limited bitcoins i mean
551 2011-01-31 18:20:00 <newsham> BTC would be nice for some micro-payment markets if they could somehow develop.
552 2011-01-31 18:20:13 <slush> newsham: yes, this is chicken-and-egg problem. Nobody will rely on bitcoins until there will be customers. So all businesses need also $$$ gateways.
553 2011-01-31 18:20:18 <citizen> there are more dollars created each second than there are total bitcoins in existance
554 2011-01-31 18:20:21 <newsham> ie. would be nice to tip helpful IRC people on programing channels with BTC ...
555 2011-01-31 18:20:57 <slush> Personally I don't think that bitcoin can live on tipping
556 2011-01-31 18:21:02 <newsham> citizen: so what?  its not the number of dollars or btc that matters.  its their purchasing power.
557 2011-01-31 18:21:28 <citizen> those things are related though
558 2011-01-31 18:21:39 <newsham> slush: no, but it is an ideal fit for btc i think..  easy to xfer, very divisible.
559 2011-01-31 18:21:46 <citizen> since bitcoins are limited and not inflatable like the dollar
560 2011-01-31 18:21:50 <citizen> their purchasing power will be greater
561 2011-01-31 18:21:50 <newsham> ie. you can give out lots of fractional pennies to lots of helpful people
562 2011-01-31 18:22:20 <slush> newsham: but until I can go to first market and buy bitcoins for almost zero fee, there is no real advantage for people who don't have bitcoins yet
563 2011-01-31 18:22:33 <newsham> citizen: therein lies a potential problem.  btc deflation.
564 2011-01-31 18:23:05 <newsham> slush: I paid 2% to acquire btc from $
565 2011-01-31 18:23:24 <slush> newsham: which way?
566 2011-01-31 18:23:29 <newsham> not to mention a long delay (which was a much bigger tax than 2% given the deflation on btc)
567 2011-01-31 18:23:32 <slush> newsham: I paid almost 10%
568 2011-01-31 18:23:50 <slush> CZK -> EUR -> USD -> BTC
569 2011-01-31 18:24:05 <slush> + paypal fee
570 2011-01-31 18:24:10 <newsham> during the delay BTC went from $35 to nearly $50
571 2011-01-31 18:24:17 <newsham> er $.35 -> $.50
572 2011-01-31 18:24:42 <sipa> wow 7000 BTC traded at 0.5
573 2011-01-31 18:25:22 <bitanarchy> sipa,  where can you see that?
574 2011-01-31 18:25:26 <newsham> deflation is going to make spending bTC unattractive.
575 2011-01-31 18:25:36 <newsham> why spend BTC today when it will be worth 2x as much in 6mos?
576 2011-01-31 18:25:51 <slush> with this speed, we cross over 0.51 soon :)
577 2011-01-31 18:26:04 <nanotube> bitanarchy: #bitcoin-market
578 2011-01-31 18:26:16 <slush> newsham: that's the reason why I'm using btc for savings now :)
579 2011-01-31 18:26:43 <bitanarchy> nanotube: where are those trades made?
580 2011-01-31 18:26:53 <nanotube> mtgox
581 2011-01-31 18:27:13 <nanotube> see topic on chan ;)
582 2011-01-31 18:27:18 <midnightmagic> ;;bc,stats
583 2011-01-31 18:27:20 <gribble> Current Blocks: 105534 | Current Difficulty: 22012.4941572 | Next Difficulty At Block: 106847 | Next Difficulty In: 1313 blocks | Next Difficulty In About: 1 week, 1 day, 8 hours, 13 minutes, and 57 seconds | Next Difficulty Estimate: 24071.59296499
584 2011-01-31 18:27:35 <TD> the focus on mining isn't so bad
585 2011-01-31 18:27:43 <TD> actually i'm happy to see such great progress
586 2011-01-31 18:28:04 <TD> one of the most common objections i've seen to bitcoin is "what about botnets" and lots of gigahashes/sec is a good answer to that
587 2011-01-31 18:28:15 <gribble> Error: "bc;mtgox" is not a valid command.
588 2011-01-31 18:28:15 <molecular> ;;bc;mtgox
589 2011-01-31 18:28:21 <newsham> Mon Jan 31 09:24:01 2011: 2000.00 @ 0.50000
590 2011-01-31 18:28:23 <molecular> we hit the big 0.51
591 2011-01-31 18:28:23 <sipa> ;;bc,mtgox
592 2011-01-31 18:28:23 <TD> i think we'll see more discussion of real economic activity over the course of this year as more basic infrastructure becomes available
593 2011-01-31 18:28:24 <gribble> {"ticker":{"high":0.51,"low":0.465,"vol":20720,"buy":0.4701,"sell":0.51,"last":0.51}}
594 2011-01-31 18:28:43 <TD> wouldn't surprise me if there are some bubbles along the way though
595 2011-01-31 18:29:00 <bitanarchy> There are hardly any bitcoins offered at bitcoin-central
596 2011-01-31 18:29:35 <sipa> chicken-and-egg problem
597 2011-01-31 18:29:38 <slush> well, first touch of .51 monster
598 2011-01-31 18:30:53 <TD> i'm hoping in 2011 we'll see:
599 2011-01-31 18:30:54 <TD> - mobile clients with NFC support
600 2011-01-31 18:31:00 <TD> - an exchanger that is accepting credit cards again
601 2011-01-31 18:31:10 <TD> - somebody set up a mixer service
602 2011-01-31 18:31:22 <TD> if those three are done by EOY 2011 it'll have been a great years progress for bitcoin, i think
603 2011-01-31 18:31:23 <sipa> - a second full bitcoin client implementation
604 2011-01-31 18:31:27 <TD> but they're all huge clients
605 2011-01-31 18:31:30 <TD> yes. that would also be great.
606 2011-01-31 18:31:41 <midnightmagic> crap my btc sold before i could modify the price!
607 2011-01-31 18:31:51 <TD> i dunno if we'll see a full 1:1 client as satoshis code has lots of half finished and unused features
608 2011-01-31 18:31:52 <midnightmagic> why didn't you people tell me before it happened!
609 2011-01-31 18:32:01 <TD> but one that's basically interoperable ....
610 2011-01-31 18:32:34 <sipa> there is a difference between requirements for the protocol, and conventions currently enforced by the default client
611 2011-01-31 18:33:51 <midnightmagic> holy crap the 20k sell is eroding
612 2011-01-31 18:34:46 <newsham> Mon Jan 31 09:27:47 2011: 673.28 @ 0.51000
613 2011-01-31 18:35:26 <newsham> mixer service?
614 2011-01-31 18:35:39 <TD> for anonymity
615 2011-01-31 18:36:05 <newsham> put money in, rolls up a big TX with many other people, get your money back out washed and pressed?
616 2011-01-31 18:36:07 <slush> isn't proposal of mixing on client enough?
617 2011-01-31 18:36:10 <TD> pretty much
618 2011-01-31 18:36:37 <slush> I preffer that before sending to some SPOF which need to be trustworthy
619 2011-01-31 18:36:38 <TD> slush: you need to mix up coins with lots of other transactions from other people. otherwise graph analysis can be a problem
620 2011-01-31 18:37:11 <slush> basically you don't mix coins when you don't have other coins
621 2011-01-31 18:37:43 <newsham> bitcoin.wash.and.fold.com ?
622 2011-01-31 18:38:13 <TD> i think a reasonably trustworthy mixer service would need to set up deals with big mining consortiums
623 2011-01-31 18:38:17 <slush> but if you have enough funds, your single bitcoin client can act as many other people
624 2011-01-31 18:38:26 <TD> like slush and ArtForz for instance :) that way you can ensure a constant flow of coins through the service
625 2011-01-31 18:38:37 <bitanarchy> newsham: why would you want to wash your money over bitcoin instead of using LR?
626 2011-01-31 18:38:46 <TD> providing cover for actual anonymization
627 2011-01-31 18:39:37 <TD> slush: not really. let's say you receive 100 BTC and want to anonymize that so you can spend it all on porn without anyone being able to link those payments back to you
628 2011-01-31 18:40:08 <TD> if you have a wallet full of anonymized coins, no problem, just don't spend those 100 BTC on porn
629 2011-01-31 18:40:30 <TD> the need for a mixer comes from people who don't already have tons of anonymized coins
630 2011-01-31 18:40:50 <ArtForzZz> foly huck
631 2011-01-31 18:41:04 <TD> you can split that 100 BTC up to lots of addresses you own but nobody else knows you own .... but then when they are all joined together again to make a purchase it's pretty clear what was happening
632 2011-01-31 18:41:31 <ArtForzZz> there goes 0.51
633 2011-01-31 18:41:37 <sipa> 0.52
634 2011-01-31 18:41:49 <TD> bubble!
635 2011-01-31 18:41:59 <slush> TD: you're right, nobody can pay the exact amount with coins which he anonymized before few minutes :)
636 2011-01-31 18:42:22 <TD> unless he uses a mixing service in which coins flow through it every few seconds
637 2011-01-31 18:42:59 <devon_hillard> have you tried overclocking your sandy bridge so far?
638 2011-01-31 18:43:06 <slush> but if you have enough funds (as I proposed before), your client can continuously mix them in your wallet, so graph of payments will be very large
639 2011-01-31 18:43:32 <slush> 15000 bitcoins traded at 0.51?
640 2011-01-31 18:43:37 <ArtForzZz> well, if all you want is plausible deniability self-mixing works
641 2011-01-31 18:43:53 <ArtForzZz> yep
642 2011-01-31 18:43:56 <ArtForzZz> more like 16500
643 2011-01-31 18:44:02 <slush> really looks like somebody is doing fun :)
644 2011-01-31 18:45:34 <midnightmagic> i am curious whether the 20,100 btc sell @ 0.51 was to test money-reporting/laundering rules.
645 2011-01-31 18:46:15 <slush> how?
646 2011-01-31 18:46:21 <slush> I mean - which rule?
647 2011-01-31 18:46:31 <ArtForzZz> yeah
648 2011-01-31 18:46:39 <midnightmagic> that's $10,050, which is $50 over the magic $10k auto-reporting limit for cash transactions in the bank
649 2011-01-31 18:46:41 <ArtForzZz> I dont see how those apply to bitcoin
650 2011-01-31 18:47:16 <midnightmagic> and neither would they apply for money transfers directly (since their nature is to be tracked)
651 2011-01-31 18:47:25 <ArtForzZz> depends a lot on country though
652 2011-01-31 18:47:27 <Sirius> midnightmagic: gox sends at most $1000 / day / user in the US
653 2011-01-31 18:47:37 <ArtForzZz> yep
654 2011-01-31 18:47:38 <slush> huh, your AML thresholds are quite high. Here we have to report transactions >=$5500
655 2011-01-31 18:47:40 <midnightmagic> I did not know that.
656 2011-01-31 18:48:08 <ArtForzZz> iirc same for mtgox transfers to EU wire
657 2011-01-31 18:48:14 <slush> you know, poor country of former Soviet Union :)
658 2011-01-31 18:48:22 <TD> where is mtgox based?
659 2011-01-31 18:48:24 <midnightmagic> $10k CAD for a deposit to the bank autogen money-laundering report forms (Canada)
660 2011-01-31 18:48:31 <ArtForzZz> costa rica
661 2011-01-31 18:50:03 <TD> midnightmagic: but 10k over what period
662 2011-01-31 18:51:10 <ArtForzZz> the question is if the bitcoin side of a bitcoin exchange even fall under AML laws
663 2011-01-31 18:51:29 <ArtForzZz> it'll probably depend on regional laws and interpretation
664 2011-01-31 18:51:47 <midnightmagic> TD: In one cash deposit.
665 2011-01-31 18:52:04 <TD> midnightmagic: that seems like a rather fragile definition
666 2011-01-31 18:52:27 <ArtForzZz> yep, iirc in .de banks have to file a report on cash deposits > 5kEUR
667 2011-01-31 18:53:20 <ArtForzZz> but then in .de using cash to pay multi-kEUR stuff isn't THAT uncommon
668 2011-01-31 18:54:05 <slush> only ~2000 USD to parity
669 2011-01-31 18:54:07 <midnightmagic> TD: It is. I ask all the time wat happens if I bring in three $9999 deposits. The tellers tell me the system is smart enough to pick that up and will generate a report for investigators semi-automatically. I've always wanted to test that and see what happens.
670 2011-01-31 18:54:08 <slush> looks strange...
671 2011-01-31 18:54:33 <ArtForzZz> yeah, there are anti-smurfing detections
672 2011-01-31 18:55:06 <midnightmagic> Three $9999 deposits to three different accounts, and then refuse to identify myself.
673 2011-01-31 18:55:22 <ArtForzZz> and they're about as useful as a pet rock
674 2011-01-31 18:55:30 <midnightmagic> =]
675 2011-01-31 18:55:43 <Sirius> ArtForzZz: no limit for EU transfers
676 2011-01-31 18:55:56 <ArtForzZz> really?
677 2011-01-31 18:56:01 <Sirius> yeah
678 2011-01-31 18:56:04 <ArtForzZz> cool
679 2011-01-31 18:56:15 <midnightmagic> No limit for cross-border Canada->US money transfers either.
680 2011-01-31 18:56:39 <midnightmagic> "A friend of mine" just transferred $50k to California and nobody cared.
681 2011-01-31 18:57:29 <ArtForzZz> well, no real reason to report wire transfers, they're already tracked by definition
682 2011-01-31 18:57:43 <midnightmagic> there're only 1515.75 BTC left in Ask orders on mtgox.
683 2011-01-31 18:58:21 <andrew12^droid> Wow
684 2011-01-31 18:58:26 <slush> midnightmagic + some dark pool
685 2011-01-31 18:58:35 <andrew12^droid> That too
686 2011-01-31 18:58:48 <nanotube> midnightmagic: http://bitcoincharts.com/markets/mtgoxUSD.html
687 2011-01-31 18:58:51 <nanotube> full order book here
688 2011-01-31 18:59:00 <midnightmagic> i was just about to ask that. thank you
689 2011-01-31 18:59:37 <ArtForzZz> + unknown amount of dark pool orders
690 2011-01-31 18:59:45 <midnightmagic> is there a way of querying that for ourselves? or does bitcoincharts have an in with mtgox?
691 2011-01-31 18:59:46 <slush> I don't believe somebody really bought those 15k BTC on 0.51
692 2011-01-31 18:59:49 <slush> it smells
693 2011-01-31 19:00:06 <nanotube> slush: well, at worst they lost the 1.3% on mtgox fees.
694 2011-01-31 19:00:25 <ArtForzZz> I doubt mtgox cares either way :P
695 2011-01-31 19:00:51 <midnightmagic> plus, if my orders are snapped up alongside it, I don't care either.
696 2011-01-31 19:01:04 <midnightmagic> and they were before i could cancel them and relist. :)
697 2011-01-31 19:02:48 <TD> i
698 2011-01-31 19:02:55 <TD> would LOVE to know who bought all those coins
699 2011-01-31 19:03:06 <slush> any advice to nice Windows IRC client?
700 2011-01-31 19:03:14 <slush> STill using XMPP transport...
701 2011-01-31 19:03:32 <bitanarchy> chatzilla
702 2011-01-31 19:03:43 <midnightmagic> you don't like mirc?
703 2011-01-31 19:03:51 <TD> webchat.freenode.net ;)
704 2011-01-31 19:03:52 <slush> I don't know mirc ;)
705 2011-01-31 19:04:11 <midnightmagic> mirc is pretty badass. or if you want open-source, xchat is built for windows too.
706 2011-01-31 19:04:21 <midnightmagic> mirc has been around forever.
707 2011-01-31 19:09:47 <slush> mirc is weird
708 2011-01-31 19:12:09 <midnightmagic> it is, but you get used to it. its way of doing things is really fast once you're used to it.
709 2011-01-31 19:13:45 <slush> well, XMPP transport is quite good, i'm only missing list of rooms and possibility to write user outside current room
710 2011-01-31 19:13:59 <slush> But I think I canl live with it
711 2011-01-31 19:14:02 <slush> *can
712 2011-01-31 19:23:35 <Vladimir> lovely market action today http://www.taters.net/cgi-bin/btc/matrix.pl?axisinc=0.01
713 2011-01-31 19:25:09 <jgarzik> wow
714 2011-01-31 19:26:10 <Vladimir> that 16k ask is gone
715 2011-01-31 19:26:19 <ArtForzZz> yup, in a single trade
716 2011-01-31 19:27:45 <bittertea> midnightmagic: Here's the mtgox trade API: http://mtgox.com/support/tradeAPI
717 2011-01-31 19:28:11 <ArtForzZz> well, either someone really really wanted some bitcoins or traded with himself and paid mtgox $110 in fees for the pleasure
718 2011-01-31 19:28:49 <bittertea> Why would you do that?
719 2011-01-31 19:28:53 <slush> ArtForzZz: Don't forget that crossing this barrier changed mind of many people. Another words - you can move the market ONLY for 110$
720 2011-01-31 19:29:03 <ArtForzZz> dunno, moving the market?
721 2011-01-31 19:29:40 <ArtForzZz> point is, I'm pretty certain plenty asks below that were perfectly normal trades
722 2011-01-31 19:30:28 <slush> probably yes
723 2011-01-31 19:30:44 <Vladimir> yea, but you risk that someone else snap your 15k ask and price run away from you
724 2011-01-31 19:30:55 <slush> but now, people think "wow, somebody push so much $$$, it would be better to buy than price go much more higher"
725 2011-01-31 19:31:23 <ArtForzZz> *shrug* people are irrational
726 2011-01-31 19:31:29 <slush> Vladimir: Pretty easy to put trade at 0.5105 dark pool
727 2011-01-31 19:31:47 <bittertea> ArtForzZz: What's the completely rational move at this point then? Wait and see? :)
728 2011-01-31 19:31:52 <molecular> holy shit: someone just bought the 16000btc from mtgox!
729 2011-01-31 19:31:55 <Vladimir> yea, gotta get more cards, but there is not stock anywhere, even in china
730 2011-01-31 19:32:14 <ArtForzZz> probably because 5970 is pretty much EOL
731 2011-01-31 19:32:18 <molecular> oh, you already noticed, should read first, then write ;)
732 2011-01-31 19:32:19 <Vladimir> wait 3 years, than see
733 2011-01-31 19:32:38 <ArtForzZz> 6990 should come out in 2 weeks or so
734 2011-01-31 19:32:56 <Vladimir> after chinese new year probably
735 2011-01-31 19:33:02 <ArtForzZz> yea
736 2011-01-31 19:33:04 <ArtForzZz> hopefully also at ~ $600 price point
737 2011-01-31 19:33:19 <Vladimir> are 6990 going to need 2x8pin power?
738 2011-01-31 19:33:22 <ArtForzZz> nope
739 2011-01-31 19:33:36 <ArtForzZz> 6+8pin and 300W max like 5970
740 2011-01-31 19:33:44 <Vladimir> ok great
741 2011-01-31 19:34:08 <ArtForzZz> theres already photos of AMD higherups showing off what looks like final cards
742 2011-01-31 19:34:31 <Vladimir> yea saw that
743 2011-01-31 19:34:38 <davout> hello all
744 2011-01-31 19:35:40 <ArtForzZz> I guess it'll be slower than 5970 for mining
745 2011-01-31 19:35:46 <jgarzik> gavinandresen: I won't be able to do it for this release, but Fedora has mingw libs built for everything bitcoin needs except wxwindows
746 2011-01-31 19:36:23 <jgarzik> gavinandresen: my goal for autotools is building bitcoin GUI on mingw
747 2011-01-31 19:36:43 <jgarzik> if it can do that, all other cases (linux CLI, windows CLI, linux GUI) are easier
748 2011-01-31 19:36:59 <ArtForzZz> 6950/70 isn't quite as efficient as 5850/70 for mining, and mining is pushing the cards a lot harder than 3D
749 2011-01-31 19:37:23 <ArtForzZz> so my guess is 6990 will have 775-800 clock and the 300W limit enforced by powertune
750 2011-01-31 19:37:52 <ArtForzZz> should be 30%+ faster than 5970 for gaming, but the power limit will affect mining pretty badly
751 2011-01-31 19:40:16 <bittertea> ArtForzZz: I thought I heard something about you using some sort of custom hashing processor. Is that working out for you?
752 2011-01-31 19:40:56 <ArtForzZz> first batch of chips should ship this or next week
753 2011-01-31 19:41:35 <bittertea> It seems that would be the natural way to go if Bitcoin takes off, a handful of dedicated devices.
754 2011-01-31 19:41:42 <ArtForzZz> yup
755 2011-01-31 19:42:22 <bittertea> Are you a CS guy, did you do the design of the processor yourself?
756 2011-01-31 19:42:29 <ArtForzZz> EE guy
757 2011-01-31 19:42:40 <bittertea> Ah, yeah, that's probably more applicable.
758 2011-01-31 19:43:56 <molecular> I can make my gpus a lot hotter now, with 1.1V ;)
759 2011-01-31 19:44:39 <ArtForzZz> did most of the HDL myself, had some help from mfg engineer for perf tuning
760 2011-01-31 19:45:10 <bittertea> Very nice... once you get the chips and have been using them for a bit, have you thought about licensing the designs?
761 2011-01-31 19:45:37 <bittertea> I was thinking that there would probably be a group of people willing to pool some money in order to invest in such an idea.
762 2011-01-31 19:45:50 <ArtForzZz> I'll probably sell finished modules
763 2011-01-31 19:46:01 <molecular> yay!
764 2011-01-31 19:46:18 <molecular> so one wont need a toaster to do the soldering?
765 2011-01-31 19:46:26 <ArtForzZz> can't comment on perf and efficiency yet, have to wait for actual silicon
766 2011-01-31 19:46:59 <bittertea> Interesting... any idea what sort of price range is likely?
767 2011-01-31 19:47:03 <ArtForzZz> but it should be on the order of 8-12x more power efficient than 5970s
768 2011-01-31 19:48:29 <Vladimir> what about the size?
769 2011-01-31 19:48:55 <ArtForzZz> hopefully 6.4Ghps in a 1U at <400W
770 2011-01-31 19:49:08 <Vladimir> nice
771 2011-01-31 19:49:28 <ArtForzZz> I could pack em tighter, but cooling will already be a major headache
772 2011-01-31 19:49:46 <bittertea> That's pretty exciting.
773 2011-01-31 19:50:14 <bittertea> Here I am wishing my coffee was here already
774 2011-01-31 19:50:23 <davout> ArtForzZz: how much will you sell those ?
775 2011-01-31 19:51:14 <ArtForzZz> I'll have to see how cheap I can get a mass-producable design
776 2011-01-31 19:51:27 <davout> makes sense
777 2011-01-31 19:51:33 <ArtForzZz> my current design is a massively overbuilt 1-off for bringup
778 2011-01-31 19:52:06 <TD> hmm
779 2011-01-31 19:52:18 <TD> i'd be willing to buy some units if you're going to sell them and there isn't too much self assembly required
780 2011-01-31 19:52:59 <TD> depends on the price i guess
781 2011-01-31 19:53:02 <ArtForzZz> = 6 layer PCBs, massively oversized VRMs, digital voltage/current readout, 2 temp sensors per chip, ...
782 2011-01-31 19:53:54 <TD> i've been thinking lately about the generalized BitX concept. in particular whether bitcoin/bitx could have application to digital voting
783 2011-01-31 19:54:19 <TD> to avoid the need to trust a central voting authority to tally the votes correcty
784 2011-01-31 19:55:17 <TD> i've been working on a design for a delegatable voting system, in which every decision facing a people is put to the vote (eg every bill before parliament) and automatic category based delegation is used to make it scale
785 2011-01-31 19:55:34 <ArtForzZz> prototype comes out to about 1500 EUR/Gh
786 2011-01-31 19:55:45 <TD> a block chain could encode peoples votes and their delegations to allow the results of the vote to be publically tallied by anyone
787 2011-01-31 19:55:59 <TD> ArtForzZz: so 9k euro for the 6Gh 1u?
788 2011-01-31 19:56:29 <TD> ArtForzZz: how much could you realistically lower it to, assuming small production runs?
789 2011-01-31 19:56:42 <TD> presumably unless you made hundreds of the things the cost would never really drop much below that
790 2011-01-31 19:56:48 <TD> especially given labor (unless you just ship out the parts)
791 2011-01-31 19:57:19 <ArtForzZz> <10kEUR final price should be doable
792 2011-01-31 19:58:06 <TD> how many Gh are you doing currently again?
793 2011-01-31 19:58:28 <ArtForzZz> right now? 17.8
794 2011-01-31 19:59:23 <bittertea> That's really interesting.
795 2011-01-31 19:59:25 <dirtyfilthy> does anyone here have a serialized genesis block header i could compare against?
796 2011-01-31 19:59:33 <ArtForzZz> these should add 19.2 GHps, assuming I manage to get 96 working chips out of 100
797 2011-01-31 20:00:07 <ArtForzZz> which will be a bit of a challenge given my limited assembly capabilities
798 2011-01-31 20:01:34 <molecular> lucky strike, it's toasted
799 2011-01-31 20:01:38 <TD> does anyone here understand satoshis proposal for a new merkle tree above multiple independent block chains?
800 2011-01-31 20:01:46 <TD> i'm trying to figure out exactly what he means/how it'd be implemented
801 2011-01-31 20:02:10 <dirtyfilthy> lucky strike: every 1000 cigarette is a finely rolled joint
802 2011-01-31 20:02:22 <dirtyfilthy> (i wish)
803 2011-01-31 20:03:54 <davout> dirtyfilthy: lol
804 2011-01-31 20:07:23 <nanotube> zomg .95 trade!
805 2011-01-31 20:07:23 <Vladimir> mtgox died
806 2011-01-31 20:07:46 <slush> wtf
807 2011-01-31 20:08:14 <bittertea> We've reached parity! ;)
808 2011-01-31 20:08:22 <slush> there was no dark pool trade
809 2011-01-31 20:08:35 <ArtForzZz> there was
810 2011-01-31 20:08:36 <donpdonp> heh. 0.95 at what volume? i can buy one btc at that price no problem :)
811 2011-01-31 20:08:45 <Vladimir> speed breaks in place, just like on nasdaq :-)
812 2011-01-31 20:09:09 <bittertea> Weird... my Bid is still in my open orders.
813 2011-01-31 20:09:27 <ArtForzZz> mtgx doesnt show bids/asks outside of x% of last trade
814 2011-01-31 20:09:32 <slush> DoM looks funny now
815 2011-01-31 20:09:39 <bittertea> ArtForzZz: Ah
816 2011-01-31 20:09:53 <andrew12> <+gribble> MTG|     TRADE|                         10 @ $0.95
817 2011-01-31 20:09:56 <andrew12> is this some sort of joke?
818 2011-01-31 20:09:57 <ArtForzZz> http://bitcoincharts.com/markets/mtgoxUSD.html
819 2011-01-31 20:10:04 <slush> probably
820 2011-01-31 20:10:12 <donpdonp> i spent more on lunch today :)
821 2011-01-31 20:10:17 <slush> somebody took his few bucks and put it to mtgox just for fun
822 2011-01-31 20:10:18 <nanotube> andrew12: a very expensive one.
823 2011-01-31 20:10:20 <slush> and for parity :)
824 2011-01-31 20:10:31 <andrew12> $10 isn't that expensive
825 2011-01-31 20:10:32 <andrew12> :p
826 2011-01-31 20:10:35 <ArtForzZz> 490 btc left to parity :P
827 2011-01-31 20:11:05 <x6763> dirtyfilthy: still need the serialized block header?
828 2011-01-31 20:11:11 <dirtyfilthy> yes plz
829 2011-01-31 20:11:40 <hacim> seems like that wont last
830 2011-01-31 20:11:47 <bitanarchy> I don't understand 10 @ $0.95. Are orders not filled against the lowest ask price?
831 2011-01-31 20:11:56 <ArtForzZz> what lower ask price?
832 2011-01-31 20:12:14 <hacim> bitanarchy: you are looking at the lower bid prices
833 2011-01-31 20:12:21 <hacim> bitanarchy: which are people sayiong "I'll buy at this amount"
834 2011-01-31 20:12:24 <chaord> hey guys...i'm just stepping into this conversation....is mtgox experiencing issues that anyone knows of?
835 2011-01-31 20:12:32 <ArtForzZz> not that I noticed
836 2011-01-31 20:12:36 <x6763> 0100000000000000000000000000000000000000000000000000000000000000000000003ba3edfd7a7b12b27ac72c3e67768f617fc81bc3888a51323a9fb8aa4b1e5e4a29ab5f49ffff001d1dac2b7c
837 2011-01-31 20:12:37 <mtgox> no it is fine
838 2011-01-31 20:12:44 <slush> chaord: looks like real trades
839 2011-01-31 20:12:59 <x6763> dirtyfilthy: i'm pretty sure that's what you're looking for
840 2011-01-31 20:13:13 <chaord> hmm...what happened to the order book then? everyone go dark?
841 2011-01-31 20:13:25 <Vladimir> yes, no asks
842 2011-01-31 20:13:31 <ArtForzZz> yup
843 2011-01-31 20:13:40 <jgarzik> if the order book is super-thin, I would buy 1 btc for $1, just to make parity for fun
844 2011-01-31 20:13:41 <chaord> alright then...well i guess we reached parity today! haha
845 2011-01-31 20:13:46 <ArtForzZz> lowest ask left is @ 0.94
846 2011-01-31 20:13:48 <dirtyfilthy> x6763: thanks
847 2011-01-31 20:13:55 <andrew12> 0.94001
848 2011-01-31 20:13:58 <bittertea> Someone(s) bought up all the ~.50 asks? Then someone bought 10 @ .95? Now the low bids do not display, correct?
849 2011-01-31 20:14:07 <ArtForzZz> yep
850 2011-01-31 20:14:07 <hacim> that was just added
851 2011-01-31 20:15:00 <bitanarchy> Now you can see an insane peak at http://bitcoincharts.com/charts/
852 2011-01-31 20:15:00 <Vladimir> interesting times
853 2011-01-31 20:15:08 <chaord> it's an interesting strategy....for someone to clean out all the ask prices, effectively resetting the market
854 2011-01-31 20:15:24 <andrew12> its also very expensive :P
855 2011-01-31 20:15:42 <nanotube> something dark at .85, it seems.
856 2011-01-31 20:15:50 <ArtForzZz> yeha, looks like it
857 2011-01-31 20:16:14 <andrew12> but I hope it goes low again, otherwise I don't have enough money to trade on mtgox anymore.
858 2011-01-31 20:16:24 <ArtForzZz> yep, dark ask @ 0.85
859 2011-01-31 20:16:26 <hacim> hah here comes all everyone trying to cash in with their .85 bids
860 2011-01-31 20:16:29 <slush> lol, now the first peak in November looks too small...
861 2011-01-31 20:18:16 <TD> anyone got theories on why somebody might be doing this?
862 2011-01-31 20:18:23 <Vladimir> LOL, mining contracts do not look so damn expensive right now
863 2011-01-31 20:19:42 <Sirius> cool, S3052's prediction of parity in january became true :)
864 2011-01-31 20:19:53 <Sirius> even if for a second only
865 2011-01-31 20:19:58 <ArtForzZz> what parity?
866 2011-01-31 20:19:59 <Vladimir> anyway, it probably will settle down in .5-.6 area in a few days
867 2011-01-31 20:20:06 <ArtForzZz> highest trade was 0.95
868 2011-01-31 20:20:19 <Sirius> oh, didn't notice
869 2011-01-31 20:20:21 <nanotube> mmm, asks coming down
870 2011-01-31 20:20:21 <Sirius> but almost
871 2011-01-31 20:20:25 <ArtForzZz> yup
872 2011-01-31 20:20:31 <chaord> ArtForzZz: you're right...it wasn't quite parity
873 2011-01-31 20:20:38 <hacim> the .85 asks are rolling in and not getting bought up
874 2011-01-31 20:20:38 <nanotube> people wanting to cash in while the cashing in is good. :)
875 2011-01-31 20:20:45 <chaord> close enough for government work though ;)
876 2011-01-31 20:21:34 <dirtyfilthy> x6763: my serialized block header looks the same as yours but doesn't seem to hash right
877 2011-01-31 20:22:08 <ArtForzZz> I think we'll see price going back down to .5-.55
878 2011-01-31 20:22:19 <x6763> dirtyfilthy: did you try hashing mine?
879 2011-01-31 20:22:39 <TD> hmm
880 2011-01-31 20:22:42 <dirtyfilthy> hang on, let me try
881 2011-01-31 20:22:51 <TD> it's too bad the tx structure doesn't have any flag bits
882 2011-01-31 20:23:06 <TD> maybe something can be done with the extraNonce part
883 2011-01-31 20:23:14 <TD> sorry
884 2011-01-31 20:23:23 <TD> that makes no sense
885 2011-01-31 20:24:34 <dirtyfilthy> yah, i only need block headers to make a hash right?
886 2011-01-31 20:24:40 <x6763> dirtyfilthy: yeah
887 2011-01-31 20:25:23 <jgarzik> hold
888 2011-01-31 20:25:30 <x6763> jgarzik: i was thinking the same thing
889 2011-01-31 20:25:47 <x6763> dirtyfilthy: version, previous block hash, merkle root hash, timestamp, difficulty, and nonce
890 2011-01-31 20:25:55 <x6763> dirtyfilthy: in that order
891 2011-01-31 20:26:22 <dirtyfilthy> yep
892 2011-01-31 20:29:32 <Vladimir> finally some bids can be seen
893 2011-01-31 20:30:05 <tcatm> Vladimir: bitcoincharts has all bids
894 2011-01-31 20:31:10 <jgarzik> Does anybody have a link, or useful bitcoin.org search terms, to find what satoshi's written on nLockTime ?
895 2011-01-31 20:31:21 <jgarzik> bitcoin.org search leaves a lot to be desired
896 2011-01-31 20:31:59 <TD> http://www.bitcoin.org/smf/index.php?topic=1786.0 ?
897 2011-01-31 20:32:04 <TD> "nTimeLock does the reverse.  It's an open transaction that can be replaced with new versions until the deadline.  It can't be recorded until it locks.  The highest version when the deadline hits gets recorded.  It could be used, for example, to write an escrow transaction that will automatically permanently lock and go through unless it is revoked before the deadline.  The feature isn't enabled or used yet, but the support is there so it co
898 2011-01-31 20:32:21 <jgarzik> yep, that's it.
899 2011-01-31 20:32:35 <jgarzik> davout: ^^
900 2011-01-31 20:33:57 <x6763> jgarzik: nTimeLock is also mentioned on the wiki Protocol page as lock_time: https://en.bitcoin.it/wiki/Protocol_specification#tx
901 2011-01-31 20:34:32 <davout> i'm not sure i really understand
902 2011-01-31 20:34:51 <jgarzik> davout: well, it's disabled so it doesn't matter in any case
903 2011-01-31 20:34:55 <davout> it means a transaction can basically be overwritten before N seconds/hours/whatever right ?