| --- Log opened Wed Nov 14 00:00:39 2007 |
| 00:01 | <@McMartin> | Hmm. The 6800 was what the 65xx was based on. Does it have an actual BCD mode? |
| 00:01 | <@AnAwesome> | Of course not. It would be too simple. |
| 00:01 | <@McMartin> | Hmm |
| 00:01 | <@McMartin> | Is this for a class? |
| 00:02 | <@AnAwesome> | There's only one command relating to BCD - |DAA |19|--****|X |2|Decimal Adjust Acc. |A=BCD format| |
| 00:02 | <@McMartin> | If not, and you just want printable decimal numbers, division by repeated subtraction is probably easier. |
| 00:02 | <@AnAwesome> | Yes. |
| 00:02 | <@McMartin> | ! |
| 00:02 | <@AnAwesome> | This is for class. |
| 00:02 | <@McMartin> | DAA may be the "fix this" command. |
| 00:02 | <@AnAwesome> | Please elaborate. |
| 00:02 | | * AnAwesome doesn't understand what it does. |
| 00:02 | <@McMartin> | I can't do so with full clarity, since I've never used it. |
| 00:03 | <@McMartin> | But I think it's "I have two BCD numbers, one digit in each nybble" |
| 00:03 | <@McMartin> | "I'm going to add them together with normal add." |
| 00:03 | <@McMartin> | "DAAing the result corrects the results to make it be the results of a BCD add." |
| 00:03 | <@McMartin> | But consult machine docmentation to make sure. |
| 00:03 | <@McMartin> | Especially consult it to see how it deals with the carry bit. |
| 00:04 | <@AnAwesome> | Wikipedia claims it only works after addition. |
| 00:04 | < Vornicus> | That will do. |
| 00:04 | < Vornicus> | (multiplication and others are harder.) |
| 00:05 | < Vornicus> | (indeed, I'm not entirely certain that DAA /can/ correct unambiguously) |
| 00:05 | <@McMartin> | You can do subtraction of n-digit numbers because A - B = A + (10^n - B). |
| 00:05 | <@McMartin> | Yeah, I'm used to DAA standing for Decimal Adjust After Addition, which was an 8086 instruction, I think. |
| 00:05 | <@McMartin> | Hmm, no. ASCII Adjust After Addition on the x86. |
| 00:06 | <@McMartin> | And ASCII Adjust before mult/div. |
| 00:06 | | * AnAwesome peers. 9's addition. How does one do that? |
| 00:06 | <@AnAwesome> | *9's complement |
| 00:07 | <@AnAwesome> | "The DAA (decimal adjust) instruction only worked after addition, and not subtraction. To subtract packed decimal, one had to use 9's complement followed by addition." |
| 00:07 | | * McMartin just showed it. |
| 00:07 | <@McMartin> | one byte covers 00-99, right? |
| 00:07 | <@AnAwesome> | Yes. |
| 00:07 | <@McMartin> | So, say you have 18 |
| 00:07 | <@AnAwesome> | Okay. |
| 00:07 | <@McMartin> | 9's complement of n in one byte is 100-n. |
| 00:07 | <@McMartin> | To subtract 1, add 99. |
| 00:07 | <@McMartin> | To subtract 2, add 98. |
| 00:08 | | * AnAwesome thinks this topic is too complex for him to be working out WITHOUT any sort of teaching going on by the fukkin' teacher. |
| 00:08 | <@McMartin> | 18+99 = 117 = 17 after rounding. |
| 00:09 | <@AnAwesome> | 100-n. Okay. I think I understand that. Now how does that work with BCD? I have to do some operations on these BCD numbers by hand to get their complement, right? |
| 00:10 | <@McMartin> | Yes, but it may be possible to do a simple transform to do it. |
| 00:11 | <@AnAwesome> | Yes? |
| 00:11 | <@McMartin> | I don't know what it is, though. =P |
| 00:11 | <@AnAwesome> | Damnit. |
| 00:11 | | * AnAwesome scratches head. |
| 00:11 | < Vornicus> | 0xAA - (bcd number) I think |
| 00:11 | < Vornicus> | ...no, that's wrong. |
| 00:11 | <@McMartin> | Then add one BCD? |
| 00:12 | | * Vornicus does the math. |
| 00:15 | < Vornicus> | 0x9A - bcd |
| 00:16 | < Vornicus> | Does not work on 0. |
| 00:16 | < Vornicus> | ...or multiples of 10. Hrm. |
| 00:16 | | * AnAwesome would probably rant at the way things get taught here in Nowhere, but after 2000 words of NaNo on top of everything else, words don't come easy. |
| 00:18 | <@McMartin> | Vorn: What if you subtract the BCD and then DAA it? |
| 00:18 | < Vornicus> | And if it is a multiple of 10, use 0xA0. Check bcd & 0x0F |
| 00:18 | < Vornicus> | McM: dunno what'll happen there. Knowing how most machines handle BCD, you might get some really crazy results. |
| 00:18 | < Vornicus> | Or, rather, knowing how most machines handle numbers in general. |
| 00:18 | | * Kyrre PatPats AnAwesome. |
| 00:18 | <@McMartin> | Yeah, but I'd think 9A DAAs to 00, and 4A to 50. |
| 00:19 | <@McMartin> | Since, well, 99 + 01, 49 + 01. |
| 00:20 | < Vornicus> | I still don't see how DAA can get the right answer in the first place. |
| 00:21 | <@AnAwesome> | Can't help you there. |
| 00:21 | < Vornicus> | 9 + 9 -> 0x09 + 0x09 -> 0x12 -> 12. It's ambiguous unless you know what one of the addends was. |
| 00:21 | <@McMartin> | Mmm, point. It must be caching something. |
| 00:21 | <@AnAwesome> | Perhaps it uses flags or something. |
| 00:22 | < Vornicus> | Can you tell me more about this chip's usual adding abilities? |
| 00:22 | < Vornicus> | Usual arithmetic abilities, that is? I know that it's a very small chip - what's the signature on addition? |
| 00:22 | <@AnAwesome> | Signature? |
| 00:22 | < Vornicus> | Can you tell it what registers to use? |
| 00:23 | <@AnAwesome> | It has accumulators A and B. It can operate on these two, or one of these and a memory cell. |
| 00:23 | < Vornicus> | okay. Where is the result placed? |
| 00:24 | <@AnAwesome> | An accumulator. |
| 00:24 | <@AnAwesome> | What's pretty funny - it doesn't have logical shift left. Only logical shift right. |
| 00:25 | < Vornicus> | it confuses me as to how that is possible. |
| 00:25 | <@McMartin> | I'm guessing ASL and LSR. |
| 00:25 | <@McMartin> | But there's no difference between ASL and LSL, ever, so. |
| 00:25 | <@AnAwesome> | Yeah. But it's pretty confusing. |
| 00:25 | < Vornicus> | Ah, well |
| 00:25 | < Vornicus> | Stop bitching about that. |
| 00:26 | <@AnAwesome> | Okay. |
| 00:26 | <@McMartin> | And ASR is, IIRC, a messy combination of ASL, and ROR. |
| 00:26 | < Vornicus> | You only have so many instructions, and with so few registers you need CISC, and with CISC you need lots of instructions. |
| 00:27 | <@McMartin> | Also, the 68xx and 65xx were, IIRC, based on the... PDP-6? |
| 00:27 | <@McMartin> | very CISC. |
| 00:27 | < Vornicus> | Making identical instructions is an exercise in siliness. |
| 00:27 | <@McMartin> | Well, making overridden menmonics is less silly. |
| 00:27 | <@McMartin> | You get 17,000 or so of those. |
| 00:27 | | * McMartin still didn't do it for Ophis. |
| 00:28 | <@McMartin> | Well, mostly. |
| 00:28 | < Vornicus> | Anyway back to DAA - what's it look like? Does it ask for an accumulator? |
| 00:29 | <@AnAwesome> | It operates strictly on accumulator A. |
| 00:29 | < Vornicus> | Okay. I'm betting that it needs accumulator B to be one of the addends. |
| 00:31 | <@AnAwesome> | http://en.wikipedia.org/wiki/Motorola_6800 <- Here's the article. |
| 00:31 | | * Vornicus fiddles with Excel, trying to get it to tell him how to do BCD. |
| 00:32 | | You're now known as TheWatcher[T-2] |
| 00:35 | < Vornicus> | The DAA (decimal adjust) instruction only worked after addition, and not subtraction. To subtract packed decimal, one had to use 9's complement followed by addition. |
| 00:35 | < Vornicus> | according to wiki. |
| 00:35 | <@AnAwesome> | Yeah. |
| 00:36 | <@AnAwesome> | What I would most like is an example. Unfortunately, the two blokes from 110 people forced to do this who actually worked it out somehow have so far failed to share. |
| 00:38 | | You're now known as TheWatcher[zZzZ] |
| 00:44 | | Thaqui [~Thaqui@Nightstar-12017.jetstream.xtra.co.nz] has joined #code |
| 00:44 | | mode/#code [+o Thaqui] by ChanServ |
| 00:47 | < Vornicus> | Yeah. Definitely requires B to be the other thing, otherwise it can't unambiguously adjust. |
| 00:48 | <@AnAwesome> | Urgh. |
| 00:53 | | * Vornicus eyes, tries something. |
| 00:53 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Z?] |
| 00:58 | < Vornicus> | okay, nines_complement(bcd_number): result = ~k - 0x65; return daa(result, 1) |
| 00:59 | <@AnAwesome> | I'm afraid I don't actually understand the notation. |
| 00:59 | <@AnAwesome> | "~k"? "daa(result, 1)"? |
| 01:00 | < Vornicus> | Sorry |
| 01:00 | < Vornicus> | DAA is decimal addition adjustment - given a result and one of the addends to get to the result, will correct to a proper BCD number |
| 01:00 | < Vornicus> | ~k should have been ~bcd_number |
| 01:00 | < Vornicus> | and ~ is bitwise not. |
| 01:02 | <@AnAwesome> | So to get the nine's complement of a BCD number, I have to first negate it, then subtract 65h and then use DAA on the result of it? |
| 01:02 | < Vornicus> | bitnot it - not negate it. |
| 01:02 | < Vornicus> | Yeah, but put 1 in B first, and I think that will make it work. |
| 01:03 | <@AnAwesome> | Put 1 in B. Okay. |
| 01:03 | < Vornicus> | Maybe. I am really uncertain of how the DAA works, so I am guessing. |
| 01:04 | <@AnAwesome> | Thanks anyway. |
| 01:05 | | * Kyrre dances with Vornicus. |
| 01:10 | | * Vornicus dances with serah |
| 01:10 | < Kyrre> | Oh! I just noticed what channel I was in. |
| 01:10 | < Kyrre> | No wonder there's all these code related questions and debates. |
| 01:10 | < Kyrre> | >_< |
| 01:11 | | * Kyrre goes back to lurking, with nothing constructive to add. |
| 01:11 | < Vornicus> | heh |
| 01:13 | | * AnAwesome thinks he will never understand how the ancients managed to work with BCD. |
| 01:14 | < Vornicus> | I can see the awful math. I don't know why they put up with it though |
| 01:14 | <@ToxicFrog> | So that, years from then, professors would have something to torment their students with. |
| 01:16 | <@AnAwesome> | Another thing. When I do the arithmetic on the BCD, and wind up with a carry - what do I do? Do I use the carry that's before the DAA or after the DAA? Do I just add the carry to the older byte or do I also DAA it? |
| 01:17 | <@AnAwesome> | Really. Fucking. Hate. Numbers. Greater. Than. Available. Arithmetic. Registers. |
| 01:20 | <@ToxicFrog> | It would be less of an issue if you didn't have to deal with BCD, too. |
| 01:21 | < Vornicus> | If you wind up with a carry on BCD. |
| 01:21 | < Vornicus> | Do the DAA, this should adjust the carry appropriately - results 100 to 127 give no carry for binary but a carry for BCD. |
| 01:22 | < Vornicus> | Then just add 1 to the higher byte if the BCD result would carry. |
| 01:22 | <@AnAwesome> | Okay. Righto. Do DAA, then add any carries and DAA? |
| 01:23 | < Vornicus> | Yeah. It's complicated to say the least, but that's what you get for multi-word arithmetic |
| 01:23 | | * AnAwesome finds it hard to blame himself here. ;p |
| 01:26 | | Doctor_Nick [~nick@Nightstar-23600.hsd1.fl.comcast.net] has quit [Connection reset by peer] |
| 01:28 | <@AnAwesome> | In the specific case of obtaining the complement, how do I handle carries? |
| 01:30 | < Vornicus> | The complement does not carry. |
| 01:31 | < Vornicus> | There will never be a carry when calculating the complement. |
| 01:31 | <@AnAwesome> | So I just do the algorithm for both bytes separately? |
| 01:32 | < Vornicus> | ...actually now that I look, that's wrong. |
| 01:33 | < Vornicus> | You have to do this in two passes, sorta - the first pass is to take each byte and do ~byte - 0x66 |
| 01:33 | < Vornicus> | Then you add one to the lowest byte, DAA, and carry as necessary. |
| 01:36 | <@AnAwesome> | Okay. |
| 01:36 | | * AnAwesome weeps. |
| 01:37 | <@AnAwesome> | I just know I'll have to beat someone to obtain a working copy of this problem, and then feign understanding, in order to pass. |
| 01:38 | <@AnAwesome> | Thanks for your help guys. Now I think I'll do something less painful, like bash my head mindlessly against a wall. |
| 01:53 | | * McMartin returns from really long seminar. |
| 01:53 | <@McMartin> | The Ancients used BCD because if you wanted to print out numbers in base 10, it was way simpler to do the math in decimal to begin with instead of using your nonexistent hardware divisors. |
| 01:54 | <@McMartin> | dividers, even. |
| 02:07 | | GeekSoldier|bed [~Rob@Nightstar-2720.pools.arcor-ip.net] has quit [Connection reset by peer] |
| 03:01 | < Vornicus> | A true thing. |
| --- Log closed Wed Nov 14 03:09:01 2007 |
| --- Log opened Wed Nov 14 03:09:06 2007 |
| 03:09 | | TheWatcher[zZzZ] [~chris@Nightstar-29731.dsl.in-addr.zen.co.uk] has joined #code |
| 03:09 | | Irssi: #code: Total of 15 nicks [7 ops, 0 halfops, 0 voices, 8 normal] |
| 03:09 | | mode/#code [+o TheWatcher[zZzZ]] by ChanServ |
| 03:09 | | Irssi: Join to #code was synced in 46 secs |
| 03:10 | | Attilla [~The.Attil@194.72.70.ns-11849] has quit [Connection reset by peer] |
| 03:20 | | Vornicus is now known as Darius |
| 03:53 | | Forj [~Forj@Nightstar-10789.ue.woosh.co.nz] has joined #code |
| 03:53 | | mode/#code [+o Forj] by ChanServ |
| 03:56 | | Chalcedon [~Chalcedon@Nightstar-10789.ue.woosh.co.nz] has joined #code |
| 03:56 | | mode/#code [+o Chalcedon] by ChanServ |
| 04:18 | | GeekSoldier|bed [~Rob@Nightstar-2720.pools.arcor-ip.net] has joined #code |
| 04:20 | | GeekSoldier|bed is now known as GeekSoldier|work |
| 05:28 | | ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has quit [Operation timed out] |
| 05:29 | | ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has joined #code |
| 05:29 | | mode/#code [+o ToxicFrog] by ChanServ |
| 05:29 | | AnAwesome [AnnoDomini@Nightstar-29148.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 05:29 | | Syloqs-AFH [Syloq@Admin.Nightstar.Net] has quit [Connection reset by peer] |
| 05:29 | | Netsplit DeepThought.NY.US.Nightstar.Net <-> Blargh.CA.US.Nightstar.Net quits: MyCatVerbs, DiceBot, @Thaqui |
| 05:30 | | Syloq [Syloq@Admin.Nightstar.Net] has joined #code |
| 05:31 | | Syloq is now known as Syloqs-AFH |
| 05:33 | | Netsplit over, joins: MyCatVerbs, DiceBot |
| 05:33 | | Netsplit over, joins: Thaqui |
| 05:33 | | mode/#code [+v DiceBot] by ChanServ |
| 05:33 | | AnnoDomini [AnnoDomini@Nightstar-29148.neoplus.adsl.tpnet.pl] has joined #Code |
| 05:33 | | mode/#code [+o Thaqui] by ChanServ |
| 06:57 | | Darius is now known as Vornicus |
| 06:58 | | GeekSoldier|work [~Rob@Nightstar-2720.pools.arcor-ip.net] has quit [Ping Timeout] |
| 07:02 | | GeekSoldier|work [~Rob@Nightstar-2918.pools.arcor-ip.net] has joined #code |
| 07:43 | | Attilla [~The.Attil@194.72.70.ns-11849] has joined #code |
| 07:46 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 07:46 | | mode/#code [+o gnolam] by ChanServ |
| 08:09 | | Vornicus is now known as Vornicus-Latens |
| 08:43 | | You're now known as TheWatcher |
| 10:13 | | Chalcedon [~Chalcedon@Nightstar-10789.ue.woosh.co.nz] has quit [Quit: Gone] |
| 10:13 | | Forj [~Forj@Nightstar-10789.ue.woosh.co.nz] has quit [Quit: Gone] |
| 10:57 | | Thaqui [~Thaqui@Nightstar-12017.jetstream.xtra.co.nz] has quit [Quit: This computer has gone to sleep] |
| 11:09 | | GeekSoldier|work [~Rob@Nightstar-2918.pools.arcor-ip.net] has quit [Ping Timeout] |
| 12:02 | <@gnolam> | http://www.rawimages.org/stuff/harddisk.jpg |
| 13:09 | <@EvilDarkLord> | Gnar. I forget, what is the command to check the version of your distro? |
| 13:13 | < Kyrre> | Ver? |
| 13:13 | < MyCatVerbs> | EvilDarkLord: uname -a, most commonly? |
| 13:14 | < MyCatVerbs> | EvilDarkLord: that gets you the kernel version, which most distros append their own name and version to. |
| 13:15 | < MyCatVerbs> | e.g. here I have: Linux snow.cs.bris.ac.uk 2.6.18-8.1.14.el5.centos.plus #1 SMP Sat Sep 29 09:51:55 EDT 2007 i686 athlon i386 GNU/Linux |
| 13:15 | < MyCatVerbs> | Not *quite* it, but close enough for government work. (This is CentOS 5, bleh.) |
| 13:15 | <@EvilDarkLord> | Thanks, MCV. Doesn't look like it worked in this case, unfortunately. Trying to find the RHEL version so I can install pysqlite. |
| 13:15 | < MyCatVerbs> | What's your uname? |
| 13:16 | <@EvilDarkLord> | Output of 'uname'? Linux. |
| 13:16 | < MyCatVerbs> | I think RHEL4 had a 2.6.16 or a 2.6.14 series kernel, or perhaps it was something earlier? I'm not sure. Anyway. |
| 13:16 | <@EvilDarkLord> | Full string is "Linux jonas-laptop 2.6.22-14-generic #1 SMP Sun Oct 14 23:05:12 GMT 2007 i686 GNU/Linux" |
| 13:16 | <@EvilDarkLord> | Er, oops. |
| 13:16 | < MyCatVerbs> | ... |
| 13:16 | < MyCatVerbs> | You've rolled your own kernel, haven't you? |
| 13:17 | <@EvilDarkLord> | Linux 128091-app1.www.ffmi.com 2.6.9-55.0.2.ELsmp #1 SMP Tue Jun 12 17:59:08 EDT 2007 i686 athlon i386 GNU/Linux |
| 13:17 | < MyCatVerbs> | Oh yeah, that's RHEL4. |
| 13:17 | <@EvilDarkLord> | (That's what I get for keeping the ssh window open in a tab next to my own compie) |
| 13:17 | <@EvilDarkLord> | Ah, cheers. |
| 13:17 | < MyCatVerbs> | I'm 2/3 certain that's RHEL4, we used to have that version installed here until the upgrade over the summer. |
| 13:17 | < MyCatVerbs> | My memory isn't perfect, but... uh, yeah. >> |
| 13:25 | <@EvilDarkLord> | By the way, have you poked any at Project Euler? |
| 13:25 | < MyCatVerbs> | Me? No, 'fraid not. |
| 13:26 | < MyCatVerbs> | Too busy reading webcomics. :) |
| 13:26 | < MyCatVerbs> | *googles* |
| 13:26 | < MyCatVerbs> | It sounds pretty awesome, thouhg. |
| 13:26 | < MyCatVerbs> | *though, argh. |
| 14:04 | | Reiv [~reiver@Nightstar-10025.xdsl.xnet.co.nz] has joined #Code |
| 14:04 | < Reiv> | 'scuse me, pasting links to and fro. |
| 14:04 | | Reiver [~reaverta@Admin.Nightstar.Net] has joined #Code |
| 14:05 | | mode/#code [+o Reiver] by ChanServ |
| 14:05 | <@Reiver> | Ah, so that's why I wasn't seeing you. ¬¬ |
| 14:07 | | Reiv [~reiver@Nightstar-10025.xdsl.xnet.co.nz] has quit [Quit: Laptop fall down go boom...] |
| 15:00 | | GeekSoldier|work [~Rob@Nightstar-2918.pools.arcor-ip.net] has joined #code |
| 15:10 | | GeekSoldier|work is now known as GeekSoldier |
| 15:42 | <@gnolam> | Balls. |
| 15:42 | | * jerith kicks them around a bit. |
| 15:43 | < GeekSoldier> | wow, jerith. you are a sadist. |
| 15:43 | < jerith> | What, they're not soccer balls? |
| 15:44 | < GeekSoldier> | I'm certain they are. gnolam was obviously upset about their disarrayed state, and you made them moreso. |
| 15:45 | | * jerith rushes off to band, it having been moved half an hour earlier. |
| 15:46 | < GeekSoldier> | enjoy jerith. what's up, gnolam? |
| 15:46 | <@gnolam> | AVR no worky. |
| 15:47 | < GeekSoldier> | did you try kicking it around a bit? |
| 15:47 | <@gnolam> | <Irish accent>Have you tried turning it off and on again?</Irish accent> |
| 15:48 | <@gnolam> | First, the code just Didn't Work. I have no idea why. Then I stopped being able to program the goddamned thing at all. |
| 15:48 | <@gnolam> | Tried power cycling the chip. Nothing. Tried power cycling the ICE. Still nothing. |
| 15:48 | < GeekSoldier> | fried? |
| 15:49 | <@gnolam> | Have no idea. :P |
| 15:49 | < GeekSoldier> | Balls. |
| 15:49 | <@gnolam> | In the end I just shouted some horrible curses and went home. |
| 15:49 | < GeekSoldier> | fair enough. |
| 15:50 | <@gnolam> | Thus: |
| 15:50 | <@gnolam> | BALLS. |
| 16:08 | | gnolam is now known as gnomatose |
| 16:33 | <@ToxicFrog> | AVR? |
| 16:46 | | Vornicus-Latens is now known as Vornicus |
| 17:14 | | You're now known as TheWatcher[afk] |
| 17:20 | <@gnomatose> | Yes. |
| 17:20 | | AnnoDomini [AnnoDomini@Nightstar-29148.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 17:27 | | AnnoDomini [AnnoDomini@Nightstar-29186.neoplus.adsl.tpnet.pl] has joined #Code |
| 17:27 | | mode/#code [+o AnnoDomini] by ChanServ |
| 18:31 | < MyCatVerbs> | ToxicFrog: Atmel-made microcontroller, I think. |
| 18:35 | < jerith> | MyCatVerbs: Correct. One of the nicer 8-but architectures. |
| 18:36 | < MyCatVerbs> | jerith: I didn't realise 8-bit arches came in nice flavours. |
| 18:38 | < jerith> | The AVR is nice. |
| 18:39 | < jerith> | Pics and 8051s (the other two I have experience with) aren't. |
| 18:39 | < jerith> | Although the 8051 is orders of magnitude less horrid than the PIC. |
| 18:39 | < Vornicus> | TF is making an 8-bit arch that doesn't sound too painful. |
| 18:39 | < Vornicus> | hell, it's risc. |
| 18:42 | < jerith> | So is the AVR. :-) |
| 18:46 | < Vornicus> | ...actually I think TF's is 16. sorry. |
| 18:50 | < MyCatVerbs> | jerith: eh, PICS aren't *so* bad. Well, yes, they suck, but they have a semi-reasonable excuse - Harvard architecture, for speed. >> |
| 18:51 | < jerith> | MyCatVerbs: There is no excuse for the memory paging craziness. |
| 18:51 | < jerith> | Nor the 8-level stack. |
| 18:52 | < MyCatVerbs> | jerith: I have never come across the former, but then I have never used anything bigger than the obscenely basic ones. |
| 18:52 | < MyCatVerbs> | jerith: the 8-level stack is, indeed, made of Fail(tm). |
| 18:53 | < jerith> | MyCatVerbs: You have 1k RAM, but only an 8-bit address register. |
| 18:53 | < MyCatVerbs> | jerith: oh, HORK. |
| 18:53 | < jerith> | So you have to manually set the upper two bits in a SFR. |
| 18:53 | < MyCatVerbs> | All that wasted effort for two measly damn bits? |
| 18:55 | < jerith> | And get this: The SFRs are spread out across pages. |
| 18:56 | < MyCatVerbs> | You've lost me, sorry. |
| 18:56 | < jerith> | So even if you only use 128 bits of RAM, you have to page. |
| 19:30 | | You're now known as TheWatcher |
| 19:35 | <@ToxicFrog> | Yes, the one I'm making is 16. |
| 19:35 | <@ToxicFrog> | And the prof doesn't want me to put PC in the general purpose register bank ;.; |
| 19:37 | < Vornicus> | heh |
| 19:37 | <@ToxicFrog> | (on the grounds that it makes pipelining harder, and even if we decide not to do the pipeline step, he wants us to design it as though we were going to) |
| 19:44 | | GeekSoldier is now known as GeekSoldier|bed |
| 20:18 | < MyCatVerbs> | Context switch on every single instruction! |
| 20:18 | < MyCatVerbs> | Makes pipelining really stunningly easier, since you no longer have to have *any* damn interlocks. |
| 22:15 | | gnomatose [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Reboot] |
| 22:19 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 22:19 | | mode/#code [+o gnolam] by ChanServ |
| 22:19 | | Pi [~sysop@Nightstar-24414.hsd1.wa.comcast.net] has quit [Connection reset by peer] |
| 22:20 | | Pi [~sysop@Nightstar-24414.hsd1.wa.comcast.net] has joined #code |
| 22:20 | | mode/#code [+o Pi] by ChanServ |
| 23:25 | | * McMartin notes in passing that aside from the lack of a multiplier or divider, the 65xx series wasn't too bad a set of 8-bit chips. |
| 23:25 | <@McMartin> | Carry, Zero, and Overflow all worked about the way you'd hope. |
| 23:32 | | * Vornicus still doesn't know what the difference between carry and overflow is. |
| 23:33 | <@AnnoDomini> | Overflow isn't set on shifts, perhaps? |
| 23:41 | <@McMartin> | Overflow is needed to distinguish between overflow and subtraction. |
| 23:41 | <@McMartin> | Adding $FF to $03 sets carry but not Overflow. |
| 23:43 | <@McMartin> | Adding $01 to $7F sets Overflow but not Carry. |
| 23:45 | <@ToxicFrog> | Ours sets overflow on both cases, and doesn't have carry ?? |
| 23:45 | <@ToxicFrog> | Or possibly it's the other way around. |
| 23:45 | | * ToxicFrog spent all day working on the control module and his brain isn't worky so good |
| 23:55 | <@McMartin> | I'd have to check, but Overflow involves a change or difference between carry and Bit 7 |
| --- Log closed Thu Nov 15 00:00:46 2007 |