00:14:35 |3b| [foobar@cpe-72-179-19-4.austin.res.rr.com] has joined #sbcl
00:30:50 -!- rpg [~rpg@216.243.156.16.real-time.com] has quit [Quit: rpg]
00:31:12 rpg [~rpg@216.243.156.16.real-time.com] has joined #sbcl
00:35:45 -!- Qworkescence [~quad@unaffiliated/quadrescence] has quit [Quit: Leaving]
01:31:09 -!- lichtblau [~user@port-92-195-61-79.dynamic.qsc.de] has quit [Ping timeout: 240 seconds]
02:34:10 echo-area [~user@182.92.247.2] has joined #sbcl
03:06:15 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl
04:30:40 loke [~elias@bb115-66-85-121.singnet.com.sg] has joined #sbcl
05:24:03 -!- loke [~elias@bb115-66-85-121.singnet.com.sg] has quit [Ping timeout: 252 seconds]
05:32:57 loke [~elias@bb115-66-85-121.singnet.com.sg] has joined #sbcl
05:40:29 -!- Cryotank2011 [~Cryotank2@c-24-17-62-152.hsd1.wa.comcast.net] has quit [Quit: Cryotank2011]
05:47:44 angavrilov [~angavrilo@217.71.227.181] has joined #sbcl
06:58:22 sdemarre [~serge@91.176.185.237] has joined #sbcl
07:06:17 easy-iPad [~easyipad@213.47.71.36] has joined #sbcl
07:32:25 -!- easy-iPad [~easyipad@213.47.71.36] has quit [Quit: Outta here?]
07:35:56 easy-iPad [~easyipad@213.47.71.36] has joined #sbcl
07:39:29 -!- sdemarre [~serge@91.176.185.237] has quit [Ping timeout: 240 seconds]
07:50:37 Cryotank2011 [~Cryotank2@c-24-17-62-152.hsd1.wa.comcast.net] has joined #sbcl
08:08:24 -!- easy-iPad [~easyipad@213.47.71.36] has quit [Quit: Outta here?]
08:15:25 easy-iPad [~easyipad@213.47.71.36] has joined #sbcl
08:47:52 -!- easy-iPad [~easyipad@213.47.71.36] has quit [Quit: Outta here?]
09:06:23 -!- Cryotank2011 [~Cryotank2@c-24-17-62-152.hsd1.wa.comcast.net] has quit [Quit: Cryotank2011]
09:51:57 homie``` [~levgue@xdsl-87-79-195-244.netcologne.de] has joined #sbcl
09:54:47 -!- homie`` [~levgue@xdsl-78-35-184-187.netcologne.de] has quit [Ping timeout: 252 seconds]
10:00:36 -!- deepfire [~deepfire@80.92.100.69] has quit [Read error: Connection reset by peer]
10:03:02 -!- stassats [~stassats@wikipedia/stassats] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
10:09:29 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.]
10:19:39 Blkt [~user@82.84.172.202] has joined #sbcl
10:32:38 -!- echo-area [~user@182.92.247.2] has left #sbcl
10:47:48 <Blkt> good day everyone
11:52:49 -!- homie``` [~levgue@xdsl-87-79-195-244.netcologne.de] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
12:59:03 easy-iPad [~easyipad@213.47.71.36] has joined #sbcl
13:09:12 -!- easy-iPad [~easyipad@213.47.71.36] has quit [Quit: Outta here?]
14:51:09 lichtblau [~user@port-92-195-61-79.dynamic.qsc.de] has joined #sbcl
14:51:48 nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has joined #sbcl
14:55:23 <nyef> G'morning all.
15:05:29 -!- nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has quit [Ping timeout: 240 seconds]
15:18:31 nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has joined #sbcl
15:18:49 -!- rpg [~rpg@216.243.156.16.real-time.com] has quit [Quit: rpg]
15:19:22 rpg [~rpg@216.243.156.16.real-time.com] has joined #sbcl
15:19:33 -!- rpg [~rpg@216.243.156.16.real-time.com] has quit [Client Quit]
15:36:48 <foom> C++ doesn't support pure imaginaries either; might check whether they have any answer.
15:37:48 drdo [~drdo@ling0.drdo.eu] has joined #sbcl
16:05:30 rpg [~rpg@mpls.sift.info] has joined #sbcl
16:21:50 -!- christoph_debian [~user@oteiza.siccegge.de] has quit [Remote host closed the connection]
16:26:44 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl
16:37:10 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.]
16:40:18 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl
16:41:53 <Kryztof> foom: do they support mixed complexes?  integer 0 and floating-point imaginary part?
16:45:29 christoph_debian [~user@oteiza.siccegge.de] has joined #sbcl
16:49:40 easy-iPad [~easyipad@213.47.71.36] has joined #sbcl
17:04:26 pchrist_ [~spirit@gentoo/developer/pchrist] has joined #sbcl
17:06:56 -!- pchrist [~spirit@gentoo/developer/pchrist] has quit [Ping timeout: 240 seconds]
17:12:35 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.]
17:19:14 -!- easy-iPad [~easyipad@213.47.71.36] has quit [Quit: Outta here?]
17:28:55 wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has joined #sbcl
17:31:40 <foom> nope
17:31:48 <wbooze> hullo
17:33:52 <pkhuong> integer 0 * inf is still a NaN (?)
17:34:07 <foom> well, there's no integer inf
17:34:27 <pkhuong> right. which is why the 0 will get upgraded to a float.
17:34:43 <foom> yea, and then it's float 0 * float inf.
17:38:11 <nyef> Shouldn't the zeroness win out over the infiniteness?
17:38:27 <nyef> Or is that just too naive an approach for this?
17:38:28 <foom> not in math.
17:39:19 <nyef> Wasn't the point of an algebraic zero that it had no effect in addition, and any multiplication involving a zero returns zero?
17:39:56 <pkhuong> nyef: infinity doesn't exist in that world.
17:40:01 <foom> okay, so, what's (5 * 0) / 0 ?
17:40:07 <foom> clearly the answer is 5, right?
17:40:22 <foom> (10 * 0) / 0 = 10, right?
17:40:41 <nyef> Okay, so this is outside what I'm used to dealing with.
17:41:01 <nyef> And no, division by zero is allowed only in the limit.
17:41:18 <foom> Right; division by 0 is how you get infinity.
17:41:50 <nyef> Okay, so infinity is on the order of a silly answer for a silly question?
17:42:09 <pkhuong> nyef: no... it's just that there are many infinities, and many zeros.
17:42:55 <nyef> ... Math is hard, I'm going to go back to thinking about network protocols instead.
17:43:05 <pkhuong> [lim x->0 1/x] and [lim x->0 1/x^2] both diverge "toward" infty, but one does so faster than the other.
17:45:17 <pkhuong> so [lim x->0 x^2/x] and [lim x->0 x/x^2] aren't the same, but 754 can't (and isn't designed to) represent that information.
17:45:39 Qworkescence [~quad@unaffiliated/quadrescence] has joined #sbcl
17:53:07 Quadresce [~quad@unaffiliated/quadrescence] has joined #sbcl
17:56:09 -!- Qworkescence [~quad@unaffiliated/quadrescence] has quit [Ping timeout: 240 seconds]
17:56:16 -!- Quadresce is now known as Qworkescence
18:01:42 stassats [~stassats@wikipedia/stassats] has joined #sbcl
18:16:39 -!- rpg [~rpg@mpls.sift.info] has quit [Quit: rpg]
18:17:15 <wbooze> does anyone of you use mcclim/beirc, mcclim/listener here ?
18:17:34 <stassats> sure you don't want to ask this question on #lisp?
18:17:45 rpg [~rpg@mpls.sift.info] has joined #sbcl
18:18:15 <wbooze> hrmm, ok then i'll ask there
18:18:37 <wbooze> actually there doesn't seem to be many using those.....
18:18:48 <wbooze> not even in/on #lisp
18:19:15 <stassats> this channel is for SBCL developers and SBCL development related discussions
18:19:24 <wbooze> i just can't get the folder/directory text fonts bigger.....
18:19:30 <wbooze> and the images too
18:19:33 <wbooze> i mean the icons
18:19:42 <stassats> please, transition to #lisp
18:19:46 <wbooze> well ok
18:37:17 -!- wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has quit [Remote host closed the connection]
19:07:56 milanj [~milanj_@93-86-114-40.dynamic.isp.telekom.rs] has joined #sbcl
19:41:26 sdemarre [~serge@91.176.185.237] has joined #sbcl
19:41:58 stassats` [~stassats@pppoe.178-66-94-103.dynamic.avangarddsl.ru] has joined #sbcl
19:42:01 -!- stassats` [~stassats@pppoe.178-66-94-103.dynamic.avangarddsl.ru] has quit [Changing host]
19:42:01 stassats` [~stassats@wikipedia/stassats] has joined #sbcl
19:45:40 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 260 seconds]
19:46:29 -!- angavrilov [~angavrilo@217.71.227.181] has quit [Read error: Operation timed out]
20:22:32 mensch [~mensch@c-174-62-210-182.hsd1.ma.comcast.net] has joined #sbcl
20:49:27 Cryotank2011 [~Cryotank2@c-24-17-62-152.hsd1.wa.comcast.net] has joined #sbcl
20:52:29 -!- nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has quit [Ping timeout: 240 seconds]
20:54:19 nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has joined #sbcl
21:07:21 <Kryztof> integer 0 * floating point Inf should probably be NaN, but it might be nice for pure-imaginary-represented-as-integer-0-and-1d0*i multiplying real +Inf to give #c(0 +Inf)
21:07:50 <Kryztof> (but you can't get there by multiplying real and imaginary parts separately)
21:09:30 <pkhuong> is it specified that we should have a cartesian representation?
21:12:33 <Kryztof> yes, more or less
21:12:50 <Kryztof> and that the two parts should be of the same type
21:12:52 <Kryztof> (clhs complex)
21:18:56 <foom> btw, locklessinc.com/articles/complex_multiplication
21:27:14 <pkhuong> foom: yeah. We punt and use a usually-accurate-enough inline routine.
21:28:00 <pkhuong> and Pearlmutter is right.
21:31:46 <foom> so, anyways, it sounds like #c(NaN Inf) is the right answer to the question, but that further code has to know that that's really an inf, not a NaN.
21:32:28 <pkhuong> foom: again, the silly thing is that this is more code than just a straight rotation
21:33:41 <foom> "straight rotation"?
21:33:45 <pkhuong> by i
21:34:37 <pkhuong> I'd love to compile (* x #c(0 1d0)) to just a component-wise swap and negate.
21:36:00 <foom> If it's a constant #c(0d0 1d0), couldn't you?
21:36:20 <pkhuong> That won't get Nan/Inf when x is Inf.
21:36:59 <foom> er, yea, of course.
21:37:32 <foom> theres no make-complex function, is there
21:37:42 <pkhuong> there's complex.
21:38:03 <foom> er, yea. so there you go. :)
21:38:09 <pkhuong> More generally, I can't shake off the feeling that there's something off about the way we approach FP optimisation. Reordering operations isn't always safe, but a lot of times, optimisations *improve* precision.
21:38:29 <foom> indeed, but it makes it unpredictable
21:38:32 <foom> that's why it's not allowed
21:39:17 <pkhuong> Stuff like disallowing the vectorisation of for (...) accu += vec[i]; is harmful to performance, readability and precision.
21:39:47 <pkhuong> There must be a smarter way to make things predictable.
21:41:12 <foom> You can always use -ffast-math.
21:41:28 <foom> (makes things faster, but doesn't help predictability)
21:41:39 <pkhuong> that enables more stuff that actively and reliably hinder precision.
21:42:10 <pkhuong> Clueful people like the ATLAS dev have complained about this before.
21:44:26 <pkhuong> converting for (...) vec[i] /= x; to vec[i] *= inv_x; <- sure, good enough. Converting division to seriously too few NR steps: not cool.
21:56:15 -!- milanj [~milanj_@93-86-114-40.dynamic.isp.telekom.rs] has quit [Quit: Leaving]
21:56:55 -!- sdemarre [~serge@91.176.185.237] has quit [Ping timeout: 240 seconds]
22:06:59 wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has joined #sbcl
22:10:28 -!- wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has quit [Client Quit]
22:15:30 wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has joined #sbcl
22:21:26 -!- christoph_debian [~user@oteiza.siccegge.de] has quit [Ping timeout: 260 seconds]
22:24:45 -!- wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has quit [Remote host closed the connection]
22:26:05 christoph_debian [~user@oteiza.siccegge.de] has joined #sbcl
22:28:19 -!- nyef [~nyef@c-174-63-105-188.hsd1.ma.comcast.net] has quit [Quit: G'night all.]
22:29:25 wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has joined #sbcl
22:44:14 -!- wbooze [~wbooze@xdsl-87-79-195-244.netcologne.de] has quit [Ping timeout: 252 seconds]
22:45:56 wbooze [~wbooze@xdsl-78-35-152-178.netcologne.de] has joined #sbcl
22:54:09 -!- whoops [u549@gateway/web/irccloud.com/x-kynoqvgqdgtxzyio] has quit [Read error: Connection reset by peer]
23:09:49 -!- rpg [~rpg@mpls.sift.info] has quit [Ping timeout: 240 seconds]
23:44:53 rpg [~rpg@216.243.156.16.real-time.com] has joined #sbcl
23:53:19 -!- Cryotank2011 [~Cryotank2@c-24-17-62-152.hsd1.wa.comcast.net] has quit [Quit: Cryotank2011]
23:55:38 whoops [u549@gateway/web/irccloud.com/x-xotcelflruaurvyi] has joined #sbcl