| --- Log opened Sat Mar 05 00:00:47 2011 |
| 00:17 | < ToxicFrog> | I have to restrain myself from ending every paragraph in this section of the paper with HOLY SHIT HOW COOL IS THAT |
| 00:17 | <@McMartin> | Hee |
| 00:19 | < ToxicFrog> | "Thus, when MIMD operation is required, the IMAPCAR2 can dynamically reconfigure itself by combining 4-PE groups to act as fully independent MIMD processing units.~\cite{kyo:2007} GODDAMN THAT IS AWESOME." |
| 00:24 | < ToxicFrog> | Seriously, this architecture I will be developing on over the summer is crazy. |
| 00:24 | < ToxicFrog> | It's a 16-bit RISC processor with 128 SIMD processing elements attached to it, each one with its own block of local memory. |
| 00:25 | < ToxicFrog> | However, there are interconnects between PEs that allow four PEs to combine their local memory, registers, and datapath to become equivalent to the main processor. |
| 00:26 | < ToxicFrog> | Allowing it to switch at runtime between a 16-bit single-core processor with heavy SIMD support to a 33-core processor without cache coherence. |
| 00:28 | | You're now known as TheWatcher[T-2] |
| 00:29 | < Alek> | omg |
| 00:30 | < ToxicFrog> | Or anything in between. |
| 00:32 | < ToxicFrog> | Alek: I know! |
| 00:32 | < ToxicFrog> | You see how I'm having so much trouble wiht this paper? |
| 00:34 | | You're now known as TheWatcher[zZzZ] |
| 00:35 | < Alek> | quite. |
| 00:39 | < ToxicFrog> | Check this shit out: the datapath of one PE is reused as the datapath of the PU; the ALUs of the other three are used as parts in the PU's FPU. Data and instruction caches are assembled from the IMEM blocks of two and the registers of one PE each. |
| 00:45 | < Alek> | basically, it's on the way to being a self-configuring block of CPU parts. XD |
| 00:55 | | Ed is now known as AnnoDomini |
| 00:57 | | AnnoDomini [annodomini@Nightstar-c04a5b58.adsl.tpnet.pl] has quit [[NS] Quit: Zzz.] |
| 00:57 | | Derakon[AFK] is now known as Derakon |
| 01:04 | | shade_of_cpux is now known as cpux |
| 01:16 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 01:17 | < gnolam> | ... and my new purpose in life is spreading http://www.youtube.com/watch?v=oWTWyLdGH4I across the entire radiation physics department. |
| 01:22 | | Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds] |
| 01:36 | | Reivles [orthia@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 02:03 | <@Vornicus> | that is a ridiculously interesting processor. |
| 02:04 | < FurryHelix> | Indeed. |
| 02:10 | < gnolam> | ? |
| 02:10 | < gnolam> | (But in "interesting processor" news, some locals just got grants to develop their THz graphene processor further) |
| 02:19 | | Kindamoody is now known as Kindamoody[zZz] |
| 02:32 | < ToxicFrog> | gnolam: IMAPCAR2, the processor I described just before you joined. |
| 02:32 | < ToxicFrog> | It's a 16-bit RISC processor with 128 SIMD elements, each of which has 4k of local memory. |
| 02:33 | < ToxicFrog> | Groups of four SIMD PEs can combine to form additional primary cores. |
| 02:33 | < ToxicFrog> | Thus, at runtime, it can sacrifice SIMD power to create additional execution cores, varying\\\\\\\\\\333333333333334errrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr r4444444444444444444 |
| 02:33 | < ToxicFrog> | between a single-core SIMD processor and a 33-core processor with no SIMD instructions. |
| 02:35 | < gnolam> | Ah |
| 02:35 | < gnolam> | ... with a bit of ? still remaining, mostly to that \\\\\... bit. |
| 02:36 | <@McMartin> | \o/, I suspect |
| 02:37 | < ToxicFrog> | Kitteh. |
| 02:45 | < gnolam> | Ah. Cats on keybords. |
| 02:45 | < gnolam> | *keyboards |
| 02:48 | < gnolam> | A friend of mind had cats who went into a phase of "HAVE TO WALK ON MASTER'S KEYBOARD" a couple of years ago. |
| 02:48 | < gnolam> | Now, getting an IM full of random characters and recognizing it as cat typing wasn't so bad. |
| 02:48 | < gnolam> | Realizing that you could tell /which cat it was/, however, was. |
| 02:50 | < Alek> | ... |
| 02:50 | < Alek> | wow. |
| 02:57 | <@Vornicus> | Heee |
| 03:02 | < gnolam> | (It was a bit easier than it should have been since one cat is borderline anorexic and the other probably sets off seismographs whenever it jumps down onto the floor, but still) |
| 03:04 | <@McMartin> | That is quality |
| 03:16 | | Attilla [Some.Dude@Nightstar-92c9199f.cable.virginmedia.com] has quit [Ping timeout: 121 seconds] |
| 03:26 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 06:46 | | Reivles [orthia@3CF3A5.E1CD01.36D449.95F5A5] has quit [Ping timeout: 121 seconds] |
| 06:54 | | Reivles [orthia@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 07:22 | | Kindamoody[zZz] is now known as Kindamoody |
| 07:32 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 07:58 | | * Vornicus plays spacechem, finds a slight improvement to pancakes: 53/18 |
| 08:19 | | Derakon is now known as Derakon[AFK] |
| 08:39 | | Kindamoody [Kindamoody@Nightstar-4764665d.tbcn.telia.com] has quit [Operation timed out] |
| 08:45 | <@jerith> | I grabbed the spacechem demo last night. |
| 08:45 | <@jerith> | Interesting, so far. |
| 08:51 | <@Vornicus> | It's fun to try to optimize the hell out of stuff but i pine for a better performance graph. |
| 08:58 | | Kindamoody [Kindamoody@Nightstar-4764665d.tbcn.telia.com] has joined #code |
| 08:58 | <@Vornicus> | I also wonder how he's doing verification. I mean this thing showed up on PCGamer, which is how i saw it. |
| 09:00 | | Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code |
| 09:01 | <@Vornicus> | I know N kinda screwed the pooch on that one, but I don't see any evidence of cheating in the graphs: you look at N's leaderboards and they're coated with people who got speeds that are even theoretically impossible. |
| 09:01 | <@Vornicus> | On the other hand I see spacechem's and I go "well... I can't do it, but it looks like it's possible..." |
| 09:18 | | You're now known as TheWatcher |
| 09:24 | | * Vornicus fights with Every Day Is The First Day. |
| 09:25 | <@Vornicus> | okay that /would/ give me it all in 121 but doesn't manage to triple bond the carbons. |
| 09:36 | <@Vornicus> | hrng. Okay, that /still/ requires me to move the hydrogens, because I can't get them that close... |
| 09:37 | <@Vornicus> | more bond sites would be great. |
| 09:38 | <@Vornicus> | No, I'm not seeing this. |
| 09:47 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 09:47 | | AnnoDomini [annodomini@Nightstar-c04a5b58.adsl.tpnet.pl] has joined #code |
| 09:47 | | mode/#code [+o AnnoDomini] by Reiver |
| 10:00 | <@Vornicus> | ...apparently someone made a one-reactor solution for in-place swap, which is /continuous/ insanity. |
| 10:05 | < gnolam> | ? |
| 10:06 | <@Vornicus> | Spacechem |
| 10:09 | | Vornicus is now known as Vornicus-Latens |
| 10:38 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Gone] |
| 10:53 | | Kindamoody is now known as Kindamoody|gym |
| 11:23 | | Rhamphoryncus [rhamph@C06FE3.F5723C.BE3FEB.9D4666] has quit [Client exited] |
| 11:44 | | AnnoDomini [annodomini@Nightstar-c04a5b58.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 11:46 | | AnnoDomini [annodomini@Nightstar-a7b3a800.adsl.tpnet.pl] has joined #code |
| 11:46 | | mode/#code [+o AnnoDomini] by Reiver |
| 12:37 | | Attilla [Some.Dude@Nightstar-92c9199f.cable.virginmedia.com] has joined #code |
| 12:37 | | mode/#code [+o Attilla] by Reiver |
| 13:05 | | Reiv [orthianz@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 13:05 | | Reivles [orthia@3CF3A5.E1CD01.36D449.95F5A5] has quit [Ping timeout: 121 seconds] |
| 13:06 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 13:10 | | Reivles [orthia@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 13:34 | | cpux [chatzilla@Nightstar-c978de34.dyn.optonline.net] has quit [[NS] Quit: ChatZilla 0.9.86 [Firefox 3.6.14/20110218125750]] |
| 13:39 | | Kindamoody|gym is now known as Kindamoody |
| 13:49 | | cpux [chatzilla@Nightstar-c978de34.dyn.optonline.net] has joined #code |
| 14:13 | | Reivles [orthia@3CF3A5.E1CD01.36D449.95F5A5] has quit [Ping timeout: 121 seconds] |
| 14:41 | | KiMo [Kindamoody@Nightstar-4764665d.tbcn.telia.com] has joined #code |
| 14:48 | | KiMo [Kindamoody@Nightstar-4764665d.tbcn.telia.com] has quit [Client exited] |
| 14:58 | | Caeldir is now known as Tarinaky |
| 15:24 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has quit [Client closed the connection] |
| 15:24 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 15:43 | | Reiv [orthianz@ServerAdministrator.Nightstar.Net] has quit [Client closed the connection] |
| 15:44 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 15:57 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has quit [Connection reset by peer] |
| 15:58 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 16:15 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code |
| 17:20 | < celticminstrel> | Eclipse is annoying me right now. I put the @Override annotation on a method, and it gives and error, but I know the method overrides something... |
| 17:35 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has quit [Client closed the connection] |
| 17:35 | | Reiv [orthianz@3CF3A5.E1CD01.36D449.95F5A5] has joined #code |
| 18:37 | | Derakon[AFK] is now known as Derakon |
| 21:21 | | Rhamphoryncus [rhamph@C06FE3.F5723C.BE3FEB.9D4666] has joined #code |
| 22:32 | | Kindamoody [Kindamoody@Nightstar-4764665d.tbcn.telia.com] has quit [Ping timeout: 121 seconds] |
| 22:32 | | Kindamoody [Kindamoody@Nightstar-4764665d.tbcn.telia.com] has joined #code |
| 23:07 | | Rhamphoryncus [rhamph@C06FE3.F5723C.BE3FEB.9D4666] has quit [Client exited] |
| 23:51 | | Vornicus-Latens is now known as Vornicus |
| --- Log closed Sun Mar 06 00:00:48 2011 |