--- Log opened Wed Aug 15 00:00:40 2018
00:03 -!- l_wl [~l_wl@ip-173-120-169-211.kscymo.spcsdns.net] has quit [Ping timeout: 240 seconds]
00:14 -!- l_wl_ [~l_wl@pool-173-66-166-232.washdc.fios.verizon.net] has quit []
00:14 -!- l_wl [~l_wl@pool-173-66-166-232.washdc.fios.verizon.net] has joined ##hplusroadmap
00:24 -!- l_wl is now known as lewl
00:27 -!- Meanderthal23 [~Meanderth@111.240.138.112] has joined ##hplusroadmap
00:29 -!- Meanderthal23 [~Meanderth@111.240.138.112] has quit [Remote host closed the connection]
00:58 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has joined ##hplusroadmap
01:06 -!- ebowden_ [~ebowden@unaffiliated/ebowden] has quit [Remote host closed the connection]
01:07 -!- ebowden [~ebowden@unaffiliated/ebowden] has joined ##hplusroadmap
01:19 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
01:34 -!- t4nk [t4nk@gateway/shell/panicbnc/x-khdcwcixdvdvswfg] has quit [Ping timeout: 276 seconds]
01:50 -!- janus18 [~janus@111.165.206.161] has joined ##hplusroadmap
01:52 -!- janus18 [~janus@111.165.206.161] has quit [Remote host closed the connection]
01:52 -!- t4nk [t4nk@gateway/shell/panicbnc/x-pbmvqosgxmljnjez] has joined ##hplusroadmap
01:52 -!- t4nk [t4nk@gateway/shell/panicbnc/x-pbmvqosgxmljnjez] has quit [Changing host]
01:52 -!- t4nk [t4nk@unaffiliated/t4nk] has joined ##hplusroadmap
01:52 -!- t4nk [t4nk@unaffiliated/t4nk] has quit [Changing host]
01:52 -!- t4nk [t4nk@gateway/shell/panicbnc/x-pbmvqosgxmljnjez] has joined ##hplusroadmap
02:40 < fenn> this reminds me of the old SKDB discussion of "what is a chair"
02:40 < fenn> .title http://youtu.be/R-1jhH4sbRk
02:40 < yoleaux> Functionality Preserving Shape Style Transfer (SIGGRAPH ASIA 2016) - YouTube
04:22 -!- nicedice [~nicedice@unaffiliated/nicedice] has joined ##hplusroadmap
05:25 -!- ExeciN [~nicexe@bnc.stormbit.net] has quit [Ping timeout: 240 seconds]
05:34 < kanzure> "the liver looks so real" cool video
05:44 -!- Lymia27 [~Lymia@197.41.10.41] has joined ##hplusroadmap
05:50 -!- Lymia27 [~Lymia@197.41.10.41] has quit [Ping timeout: 244 seconds]
05:50 -!- ExeciN [~nicexe@bnc.stormbit.net] has joined ##hplusroadmap
06:10 < kanzure> https://www.wired.com/story/foreshadow-intel-secure-enclave-vulnerability/amp
06:21 < kanzure> gp-write wants a wiki but they want it access-controlled. what's the point? what are they even going to put on there?
06:36 < kanzure> culture shock electroporator v0.6 https://imgur.com/gallery/t40v7j0
06:44 < kanzure> .title https://www.youtube.com/watch?v=bXM7Hx51Qb0
06:44 < yoleaux> Procedural Building Generator in Houdini - YouTube
06:45 < fenn> houdini can do some wild shit
06:45 < fenn> too bad it costs $5000/yr
06:45 < kanzure> i was looking at their site and i'm not really sure what houdini is.
06:47 < kanzure> .tw https://twitter.com/kanzure/status/1029726155736330240
06:47 < yoleaux> Holy grail of video game development is that sweet spot where an hour of a content creator's time can create an hour of content, or 10 hours of content, or 100 hours of content. Right now it's more like 1 hour of content creator time makes 0.01 hour of game content. (@kanzure)
06:48 < fenn> it's a procedural graphics system, basically you write a bunch of code by connecting boxes with lines, and there are really fast vector and volume operations like numpy
06:48 < fenn> if you've seen a movie with computer graphics in the last 20 years it's probably using houdini
06:49 < fenn> .title http://vimeo.com/267270128
06:49 < yoleaux> Houdini 16.5 Tree/Soil Wedges on Vimeo
06:50 < fenn> .title http://vimeo.com/110131273
06:50 < yoleaux> Fire and Smoke sims on Vimeo
06:51 < fenn> .title http://youtu.be/9HI8FerKr6Q
06:51 < yoleaux> a unified approach to grown structures - YouTube
06:57 -!- Dan_Bennett [~Dan_Benne@110.86.79.126] has joined ##hplusroadmap
06:58 -!- Dan_Bennett [~Dan_Benne@110.86.79.126] has quit [Remote host closed the connection]
07:05 < fenn> some of this stuff should look familiar: http://cargocollective.com/mattebb/
07:06 < fenn> matt ebb uses mostly houdini and some blender and maybe writes his own code but who can know
07:19 < fenn> i was complaining the other day about how video games spend trillions of processor cycles on graphics and hardly any is left over at the end for doing actual story-relevant simulation, like what the agents are thinking about their environment and other agents, and how that evolves over time
07:20 < fenn> dwarf fortress is such a hacky piece of crap but it's one of the very few games that actually tries to model what's going on in the world
07:20 < fenn> as a result we have "open world" games with pre-scripted storylines scattered throughout
07:21 < fenn> but if you deviate from those stories at all the whole thing falls apart
07:22 < fenn> usually this means you can't do anything that the game designers hadn't thought of in advance
07:36 < Jay_Dugger> fenn, do you know of any games that use a player's saved game files to model the world?
07:37 < Jay_Dugger> E.g., a haunted house game that uses the player's earlier saved game files to show the player earlier versions of himself as visions or ghosts?
07:38 < Jay_Dugger> Or video recordings of earlier game-play appearing as plot coupons later in the game?
07:57 < ebowden> I wonder if a console with a giant multi-chip module at its core, containing a powerful, high clock speed multi-die CPU, a powerful multi-die GPU, a specialised physics processing unit and some other chips would be good for this.
08:14 < fenn> Jay_Dugger: all i'm coming up with is racing game time trials...
08:17 < ebowden> fenn, so, modern games still put loads of graphics operations into the CPU, even with GPUs being commonplace?
08:17 < fenn> no, i meant games ought to simulate more complex phenomena on the GPU
08:18 < fenn> basically all progress in computers in the past decade has been GPU parallelization
08:18 < fenn> for some reason CPU parallelization has only just now started to increase again
08:19 < fenn> (also power consumption for mobile phones but meh)
08:20 < fenn> a single core CPU from 10 years ago is about the same performance-wise
08:20 < ebowden> What is a single core CPU from 10 years ago the same as in terms of performance?
08:21 < fenn> the same as a modern single core cpu
08:21 < ebowden> wut
08:21 < ebowden> Where can you even get modern single core CPUs?
08:21 < ebowden> What are they used for?
08:21 < fenn> running single threaded code? :P
08:22 < ebowden> What devices use them?
08:22 < fenn> if there's one thread running in your program, you're limited to a single core
08:23 < fenn> i don't consider four or eight cores to be a big improvement
08:24 < fenn> that's not even an order of magnitude (less if you take into account amdahl's law)
08:26 < ebowden> fenn, you heard the wonderful news about AMD giving Intel a market ass-raping?
08:26 < fenn> am i supposed to celebrate?
08:26 < fenn> intel seems to do a good job
08:28 < fenn> at least i haven't heard of any obviously malicious actions against their users
08:30 < fenn> also the supposed amazing advance of 32 threads doesn't seem that impressive to me, merely business as usual
08:31 < fenn> like, it's about time
08:35 < ebowden> I am interested to see what the actual competition could achieve.
08:37 < ebowden> I doubt we will see them fined large sums for anticompetitive practices again for a while yet.
08:38 < fenn> i'd like to see: reconfigurable FPGA computing, optical interconnects, highly 3D VLSI layouts, and a better theoretical software paradigm than the turing machine that respects the location of data in a computing system
08:39 < ebowden> FGPAs get a decent boost from on-die FPUs.
08:40 < fenn> a reconfigurable analog gate array would do a lot of this "TPU" stuff faster and with less energy
08:41 < fenn> neural networks and so on
08:42 < fenn> it would take a bit of an attitude adjustment to get used to the non-determinism
08:44 < ebowden> Someone made the world's first analogue compiler recently.
08:44 < fenn> lol no
08:44 < ebowden> They didn't?
08:45 < ebowden> It is not a true analogue compiler, or a true analogue compiler existed far before it?
08:46 < fenn> there were a couple decades of analog computer use that overlapped digital computers
08:46 < fenn> they still use analog computers for really fast stuff like phased array radar or winglet actuator control on fighter jets (or something like that)
08:47 < ebowden> Custom analogue hardware would also be a lot cheaper to fab.
08:47 < ebowden> fenn, lots analogue in missiles.
08:48 < ebowden> Was 130nm process node there was no point going below for analogue chips?
08:49 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:dcac:dc60:5e49:5c46] has joined ##hplusroadmap
08:54 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mxmemnmqndikqege] has joined ##hplusroadmap
08:58 -!- diogenese19 [~diogenese@139.195.10.158] has joined ##hplusroadmap
09:00 -!- diogenese19 [~diogenese@139.195.10.158] has quit [Remote host closed the connection]
09:02 < fenn> .title http://vimeo.com/229252148
09:02 < yoleaux> Procedural Realistic Tree R&D on Vimeo
09:03 < fenn> right out of art school and she does this in 2 months, best plant generator software i've ever seen
09:03 -!- zz_ka6sox [~zz_ka6sox@2a01:540:8f04:7400:21f:ceff:fe62:a551] has joined ##hplusroadmap
09:04 -!- zz_ka6sox [~zz_ka6sox@2a01:540:8f04:7400:21f:ceff:fe62:a551] has quit [Remote host closed the connection]
09:40 < Jay_Dugger> Wasn't there a reconfigurable FPGA project in the late 1990s?
09:41 < Jay_Dugger> and thank you fenn, I hadn't thought of time trials or sports games with "instant replay". Not quite what I had in mind, but that's the right idea, yes.
10:07 < ebowden> Someone made a FPAA.
10:11 < Jay_Dugger> Do you remember the name? I think it had "star" in it, but doubtless that's too vague.
10:12 < ebowden> What about the critically acclaimed game Super Hot?
10:14 < Jay_Dugger> Oh, I haven't played that. Maybe. Thank you for mentioning it.
10:14 -!- aeiousomething [~aeiousome@unaffiliated/aeiousomething] has quit [Ping timeout: 240 seconds]
10:46 -!- pOe[ [~pOe@177.48.178.237] has joined ##hplusroadmap
10:50 -!- pOe[ [~pOe@177.48.178.237] has quit [Remote host closed the connection]
10:57 -!- aeiousomething [~aeiousome@unaffiliated/aeiousomething] has joined ##hplusroadmap
11:54 -!- Xoc17 [~Xoc@121.172.193.61] has joined ##hplusroadmap
11:58 -!- Xoc17 [~Xoc@121.172.193.61] has quit [Remote host closed the connection]
12:10 -!- Typhon10 [~Typhon@213.197.71.50] has joined ##hplusroadmap
12:15 -!- Typhon10 [~Typhon@213.197.71.50] has quit [Remote host closed the connection]
13:21 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has quit [Quit: Malvolio]
13:46 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has quit [Ping timeout: 272 seconds]
13:59 -!- jqtrde_ [~jqtrde@pool-100-15-235-160.washdc.fios.verizon.net] has joined ##hplusroadmap
14:01 -!- sachy [~sachy@78.108.102.220] has joined ##hplusroadmap
14:13 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
15:19 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 240 seconds]
15:36 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
15:40 -!- jqtrde_ [~jqtrde@pool-100-15-235-160.washdc.fios.verizon.net] has quit [Ping timeout: 244 seconds]
15:41 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has joined ##hplusroadmap
15:41 -!- traumschule [~traumschu@mehl.schokokeks.org] has quit [Remote host closed the connection]
15:45 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has quit [Read error: Connection reset by peer]
15:51 -!- traumschule [~traumschu@mehl.schokokeks.org] has joined ##hplusroadmap
15:51 -!- traumschule [~traumschu@mehl.schokokeks.org] has quit [Remote host closed the connection]
16:03 -!- traumschule [~traumschu@mehl.schokokeks.org] has joined ##hplusroadmap
16:06 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
16:29 -!- sachy [~sachy@78.108.102.220] has quit [Quit: Leaving.]
17:16 -!- mal18 [~mal@185.62.188.154] has joined ##hplusroadmap
17:17 -!- mal18 [~mal@185.62.188.154] has quit [Remote host closed the connection]
17:43 -!- pheizax [~pheizax@218.85.101.206] has joined ##hplusroadmap
17:44 -!- Gurkenglas [~Gurkengla@unaffiliated/gurkenglas] has quit [Ping timeout: 240 seconds]
17:47 -!- pheizax [~pheizax@218.85.101.206] has quit [Remote host closed the connection]
17:48 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has quit [Ping timeout: 272 seconds]
18:08 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has joined ##hplusroadmap
18:24 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has quit [Ping timeout: 240 seconds]
18:27 < kanzure> hmph
19:32 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has joined ##hplusroadmap
20:27 -!- justan0theruser is now known as justanotheruser
20:38 -!- lewl [~l_wl@pool-173-66-166-232.washdc.fios.verizon.net] has quit [Remote host closed the connection]
20:49 -!- lewl [~l_wl@pool-173-66-166-232.washdc.fios.verizon.net] has joined ##hplusroadmap
20:57 -!- traumschule [~traumschu@mehl.schokokeks.org] has quit [Quit: Lost terminal]
21:00 -!- traumschule [~traumschu@mehl.schokokeks.org] has joined ##hplusroadmap
21:10 -!- hehelleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
21:14 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has quit [Ping timeout: 260 seconds]
22:04 -!- Boohbah23 [~Boohbah@14.100.34.61] has joined ##hplusroadmap
22:06 -!- Boohbah23 [~Boohbah@14.100.34.61] has quit [Killed (Sigyn (Spam is off topic on freenode.))]
22:14 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mxmemnmqndikqege] has quit [Quit: Connection closed for inactivity]
22:16 -!- yashgaroth [~yashgarot@2606:6000:c308:f700:dcac:dc60:5e49:5c46] has quit [Quit: Leaving]
22:29 < nmz787> there's the PSoC chips with some reconfigurable analog stuff
22:29 < nmz787> that are recently produced
22:30 < nmz787> http://www.cypress.com/products/psoc-analog-coprocessor
22:57 -!- maaku [~maaku@173.234.25.100] has quit [Quit: ZNC - https://znc.in]
22:58 -!- maaku [~maaku@173.234.25.100] has joined ##hplusroadmap
23:28 -!- steven3 [~steven@177.125.23.207] has joined ##hplusroadmap
23:33 -!- steven3 [~steven@177.125.23.207] has quit [Ping timeout: 260 seconds]
23:35 -!- jqtrde_ [~jqtrde@pool-108-31-165-46.washdc.fios.verizon.net] has quit [Ping timeout: 240 seconds]
23:39 < maaku> nmz787: is there an analog circuit scene? is there tooling, infrastructure for making such circuits?
23:40 < maaku> because there's an interesting parallel between analog circuits and arithmetic circuits, for which there are efficient zero-knowledge proofs
23:42 < maaku> ugh why is everyone so enamored with The Three Body Problem?
23:44 < nmz787> maaku: not too much 'scene' level outside audiophile and synthesizer/guitar-effects-pedal types
23:45 < nmz787> maaku: I recall a figure something like there are 5000 expert-level analog engineers in the world, while there are millions of digital experts with similar depth of knowledge/domain-mastery
23:46 < nmz787> maaku: but maybe sam zeloof would know better... he has at least made transistors at home
23:46 -!- armyriad29 [~armyriad@119.207.115.143] has joined ##hplusroadmap
23:48 -!- armyriad29 [~armyriad@119.207.115.143] has quit [K-Lined]
23:52 < maaku> I was more thinking if there was like high-level DSP language to circuit compiler, some of the underlying tooling for building arithmetic circuits from high-level descriptions might be reusable
23:53 < maaku> for compiling arithmetic circuits for non-analog stuff, like rangeproofs
23:53 < nmz787> hmm, maybe but I haven't come across anything that immediately strikes me... my 'maybe' also is leaning towards ultra expensive licensed software like ansys or something
23:54 < maaku> yeah
23:55 < nmz787> but from what I know it's much more human driven, then validated in lots of different pieces and tools with simulation, then iterations of actual builds
23:55 < nmz787> (than digital design)
23:55 < nmz787> seems reasonable that you could code a synthesizer based on some language
--- Log closed Thu Aug 16 00:00:41 2018