| 02:03.24 | *** join/#edev prpplague (n=dave@69.73.213.167) |
| 02:41.48 | aundry | what the fuck |
| 02:41.55 | aundry | ISO wants MONEY for their spec sheets o.0 |
| 02:49.49 | aundry | does anybody here have an ANSI/ISO subscription? |
| 03:21.43 | joe_bleau | What spec are you looking for? |
| 03:47.34 | *** join/#edev ohsix (i=ohsix@bc111111.bendcable.com) |
| 03:52.05 | *** join/#edev redblue (n=star@ppp119.109-253-207.mtl.mt.videotron.ca) |
| 04:07.28 | *** join/#edev ohsix_ (i=ohsix@bc111123.bendcable.com) |
| 04:08.57 | *** join/#edev ohsix (i=ohsix@bc111123.bendcable.com) |
| 04:56.40 | *** part/#edev joe_bleau (n=joe_blea@adsl-69-152-192-31.dsl.fyvlar.swbell.net) |
| 06:10.35 | *** join/#edev ohsix_ (i=ohsix@bbb72-0-186-149.bendbroadband.com) |
| 06:36.42 | *** join/#edev ohsix (i=ohsix@bc111222.bendcable.com) |
| 07:22.32 | *** join/#edev ohsix_ (i=ohsix@bc111222.bendcable.com) |
| 07:23.45 | *** join/#edev ohsix (i=ohsix@bc111222.bendcable.com) |
| 11:17.48 | *** join/#edev redblue (n=star@ppp149.109-253-207.mtl.mt.videotron.ca) |
| 11:32.37 | *** join/#edev prpplague^2 (n=dave@72.22.129.40) |
| 11:41.13 | *** join/#edev prpplague^3 (n=dave@69.73.209.248) |
| 11:58.49 | *** join/#edev prpplague^2 (n=dave@72.22.141.65) |
| 12:06.05 | sunrunner20 | http://webstore.ansi.org/ansidocstore/product.asp?sku=ISO+11785%3A1996 |
| 12:06.17 | sunrunner20 | give that to joe bleau if i'm not around next time he shows |
| 12:46.02 | *** join/#edev GPSFan (n=Ken@12.10.255.248) |
| 12:46.02 | *** mode/#edev [+o GPSFan] by ChanServ |
| 12:46.14 | prpplague | GPSFan: hey |
| 12:46.25 | GPSFan | prpplague: happy May... |
| 12:47.06 | prpplague | GPSFan: merry freakin monday |
| 12:47.20 | GPSFan | hehe |
| 12:50.31 | prpplague | GPSFan: you every run across anything using pwm for audio? |
| 12:53.39 | GPSFan | prpplague: you mean like a pc speaker driver? |
| 12:56.58 | prpplague | GPSFan: yea, similiar but more on the hardware side |
| 12:59.14 | prpplague | GPSFan: the pixter color aparently uses just a pwm with an external cap for audio |
| 12:59.19 | GPSFan | prpplague: no, can't say that I have, I remember a pwm driver fro the pc speaker that let you play crappy sounding audio through a 1 bit, bit-banged, i/o port, but that was a long time ago in the dark days of DOS |
| 13:00.07 | prpplague | GPSFan: yea |
| 13:02.24 | prpplague | GPSFan: seems the main reason for the low quality is that the speaker is on a standard trimer instead of a pwm |
| 13:03.02 | GPSFan | trimer.? timer maybe? |
| 13:04.57 | GPSFan | yeah a timer output would hose things up a little. but without it, it may be hard to send the proper # of transitions/second to the port due to the CPU being hogged by other stuff (low performance systems, like 5MHz 8088) |
| 13:05.30 | prpplague | GPSFan: yea |
| 13:05.54 | prpplague | GPSFan: but since most arm's have pwm and have a relatively high mhz compared to older x86's |
| 13:06.04 | prpplague | GPSFan: you can create better sound with a pwm |
| 13:08.16 | GPSFan | prpplague: correct. |
| 13:19.08 | jbevren | Ive heard impressive pwm on a 386/25 heh |
| 13:19.12 | jbevren | also, morning |
| 13:20.10 | prpplague | jbevren: oh yea? |
| 13:20.18 | prpplague | jbevren: i assume it was mono |
| 13:21.01 | jbevren | yeah, pc-speaker output |
| 13:21.11 | prpplague | interesting |
| 13:21.18 | jbevren | look at (for example) impulse tracker, which does great pwm on a 486/133 |
| 13:21.27 | jbevren | I think the 386 was running cubetracker |
| 13:21.39 | jbevren | or something like that (cubic player? it as cubed whatever it was) |
| 13:21.46 | prpplague | jbevren: impulse tracker? |
| 13:22.02 | jbevren | nod |
| 13:22.08 | jbevren | its a dos-based music tracker |
| 13:22.18 | prpplague | ahh |
| 13:22.52 | prpplague | jbevren: problem i'm looking into now is my pwm can't do exactly 44.1khz or 48khz |
| 13:23.13 | jbevren | ah |
| 13:23.24 | jbevren | software-resample to 22khz and go? :) |
| 13:23.30 | prpplague | jbevren: with my prescaler and all, i can do 44.3 or 48.02 |
| 13:23.58 | prpplague | jbevren: not sure i can get exactly 22khz either |
| 13:24.51 | prpplague | yea i can get pretty darn close to 22 |
| 13:25.19 | prpplague | 22.005 |
| 13:25.33 | jbevren | Close Enough(TM) |
| 13:25.35 | prpplague | think that would do for some basic audio |
| 13:25.40 | jbevren | youre not making a professional synthesizer :) |
| 13:25.59 | prpplague | jbevren: no, just wonder how that would effect the audio |
| 13:26.28 | jbevren | a slight pitch downward |
| 13:26.31 | jbevren | heh |
| 13:27.45 | prpplague | i wonder if i can drive a small speaker directly without any filtering |
| 13:30.53 | jbevren | at 22khz? you'd irritate the local cats and dogs :) |
| 13:30.53 | jbevren | mwahaha |
| 13:30.59 | jbevren | http://members.aol.com/MJMahon/ <- this guy PWM's 22khz on an apple2 cluster |
| 13:31.05 | prpplague | ahh thanks |
| 13:31.08 | jbevren | ( section "AppleCrate 8-Voice Synthesizer") |
| 13:31.14 | prpplague | i had heard that the apple had pwm audio |
| 13:31.17 | jbevren | that'll be in 6502 assembly source heh |
| 13:31.22 | jbevren | well |
| 13:31.25 | jbevren | all the apple can do is click :) |
| 13:31.43 | jbevren | the PC can actually generate tones on its speaker hehe |
| 13:33.55 | prpplague | ahh thanks for the page |
| 13:33.59 | prpplague | looks like good info |
| 13:34.14 | prpplague | yea i guess i need a filter to get rid of the carrier freq |
| 13:37.04 | GPSFan | prpplague: usually the speaker itself can't respond to anything near the 22khz rate, so, filtering is not much of an issue. |
| 13:41.43 | jbevren | good point |
| 14:25.46 | jbevren | yawn |
| 14:52.53 | *** join/#edev EvilDevil_ (n=evildevi@p54A6F967.dip.t-dialin.net) |
| 14:59.17 | prpplague | GPSFan: hmm, ok, wonder why so many examples of using pwm use low-pass filter? |
| 14:59.28 | jbevren | because pwm has high-pass noise :0 |
| 14:59.30 | jbevren | <PROTECTED> |
| 14:59.33 | jbevren | er high-band noise that is |
| 14:59.44 | jbevren | the LPF is used to strip this noise off |
| 15:00.02 | prpplague | ahh ok, miss understanding then |
| 15:00.15 | jbevren | if youre driving a powerful driver also, you can damage the driving element by sending it frequencies it cant reproduce |
| 15:00.17 | prpplague | i was thinking the low-pass was to remove the carrier freq |
| 15:00.24 | jbevren | it is :) |
| 15:00.31 | jbevren | the carrier freq is the high-band noise I'm speaking of |
| 15:00.38 | prpplague | oh ok gotcha |
| 15:01.01 | sunrunner20 | i've got an awsome idea that i'm hopeing to be able to work ^_^ |
| 15:01.20 | prpplague | oh? |
| 15:01.22 | sunrunner20 | a triangulation systme that uses pet ID RFID takes to track your pets |
| 15:01.48 | davidc__ | sunrunner20: over what distance? |
| 15:01.49 | jbevren | wouldnt work. I refuse to normalize the use of RFID to track living beings |
| 15:01.58 | davidc__ | and RFID is fuck hard to get the range up |
| 15:02.15 | davidc__ | I spent a good long time banging my head against a wall to get 125khz RFID to read over 2cm |
| 15:02.17 | sunrunner20 | jbevren:its already being done |
| 15:02.23 | jbevren | that doesnt mean I have to support it |
| 15:02.34 | jbevren | next thing you know, they'll require humans to have rfid |
| 15:02.47 | sunrunner20 | i'd prefer it be optional |
| 15:03.06 | jbevren | it'll become as optional as registering for selective service |
| 15:03.24 | sunrunner20 | i like the idea that my cat can have(i want it to, if i have anymore outdor pets) so if it gets captured they can return it to me |
| 15:03.37 | sunrunner20 | and it doesn't have to be rfid |
| 15:03.47 | sunrunner20 | i'm thinking if i can't get rfid method to work i'll use a special collar |
| 15:04.00 | jbevren | hehe |
| 15:04.07 | jbevren | see, we have these things, called name tags |
| 15:04.08 | jbevren | :P |
| 15:04.09 | sunrunner20 | with flexable solar panels as power |
| 15:04.13 | jbevren | they hang off a collar |
| 15:04.27 | sunrunner20 | yea but they don't tell you your cats in the closet somewhere where you can't see it |
| 15:04.31 | jbevren | theyre nice cause little suzy wont need a rfid reader to tell where your pet should go when she finds it |
| 15:04.48 | sunrunner20 | i know |
| 15:04.54 | sunrunner20 | ours have tags even though they are insid |
| 15:05.06 | jbevren | besides, if your pet leaves and doesnt come back, you probably didnt deserve the pet :) |
| 15:05.42 | jbevren | take for example my brother's cat |
| 15:05.49 | jbevren | he couldnt have a pet at his rental |
| 15:05.57 | jbevren | so he took the cat to a friend's farm where he was well cared for |
| 15:06.10 | jbevren | the cat went missing, and two weeks later showed up meowing at my brother's rental |
| 15:06.16 | jbevren | that's a deserved pet. |
| 15:08.01 | sunrunner20 | yea |
| 15:08.04 | sunrunner20 | but some times people take them |
| 15:08.06 | sunrunner20 | besides |
| 15:08.15 | sunrunner20 | we've lost our cats INSIDE our own house |
| 15:08.20 | davidc__ | sunrunner20: oh, you're talking about triangularing across a wide area with RFID? |
| 15:08.30 | davidc__ | sunrunner20: in any case - RFID is probably not the technology you want |
| 15:08.38 | sunrunner20 | i figured |
| 15:08.38 | davidc__ | RFID is passive - it needs to be excited somehow |
| 15:09.01 | sunrunner20 | but i'd have to look at the ISO specs for the rfid method that is implanted in cats |
| 15:09.08 | sunrunner20 | i know rfid is passive |
| 15:09.11 | sunrunner20 | which is WHY i like it |
| 15:09.15 | sunrunner20 | no batteries to be replaced |
| 15:09.23 | davidc__ | sunrunner20: the problem is is how are you planning to excite the tag over large distances :P |
| 15:09.28 | davidc__ | you'd need coils the size of a city |
| 15:09.32 | sunrunner20 | lol |
| 15:09.37 | davidc__ | and then anything metal would be wrenched around |
| 15:09.49 | *** mode/#edev [+o prpplague] by ChanServ |
| 15:09.52 | davidc__ | not to mention the problem of distinguishing between all the pets that are in range |
| 15:10.00 | sunrunner20 | rfid is.. |
| 15:10.02 | sunrunner20 | RFID... |
| 15:10.16 | sunrunner20 | would time date teh signal and then id it |
| 15:10.31 | davidc__ | sunrunner20: the problem is that there isn't a signal |
| 15:10.37 | davidc__ | its a modulation imposed on a carrier |
| 15:10.59 | davidc__ | and the tags I'm familiar with just keep repeating their modulation - which means that all tags will be spitting out their modulation at once |
| 15:11.05 | davidc__ | which means you'll be seeing garbage |
| 15:11.52 | sunrunner20 | and all this info is inside the iso specs |
| 15:11.57 | sunrunner20 | which piss me off becuase they cost money |
| 15:12.05 | sunrunner20 | but |
| 15:12.10 | davidc__ | sunrunner20: just see which frequency it runs at |
| 15:12.11 | sunrunner20 | i probally have access indirectly |
| 15:12.12 | davidc__ | surely thats available |
| 15:12.22 | sunrunner20 | dad works for xilinx |
| 15:12.25 | davidc__ | if its 125khz RFID - good luck actually exciting that thing |
| 15:12.31 | agaffney | the whole idea of a "standard" that costs money is rediculous |
| 15:12.33 | davidc__ | sunrunner20: ok, thats cool - xilinx makes awesome stuff |
| 15:12.39 | sunrunner20 | shurly xilinx has a iso/ansi subscription |
| 15:12.49 | davidc__ | sunrunner20: tell him to get me some free parts :D |
| 15:12.54 | sunrunner20 | i've got access to an FX12 demo board :p |
| 15:13.06 | jbevren | I'll accept a virtex-4 board |
| 15:13.10 | jbevren | :D |
| 15:13.14 | sunrunner20 | and i think a FX40 too |
| 15:13.16 | davidc__ | sunrunner20: daaamn yoooouu... |
| 15:13.20 | sunrunner20 | lol |
| 15:13.26 | sunrunner20 | we've got a crapload of stuff |
| 15:13.28 | davidc__ | I can only afford little spartan 3's :( |
| 15:13.30 | sunrunner20 | even BEFORE he worked for xilinx |
| 15:13.46 | jbevren | oh I see that xilinx finally released the 3e starter board |
| 15:13.48 | sunrunner20 | davidc__: xilinx is pretty liberal with their 3E thumb boards |
| 15:13.50 | jbevren | 9 months late |
| 15:13.51 | davidc__ | I'm trying to persuade the local xilinx rep to give me a V4 board for my startup.... so they get the design win rather than someone |
| 15:14.04 | davidc__ | sunrunner20: ya, I know - but for what I want to do - I need a V4 :P |
| 15:14.06 | sunrunner20 | davidc__: where are you located? |
| 15:14.14 | davidc__ | jbevren: already bought one ;) |
| 15:14.18 | davidc__ | sunrunner20: Vancouver, BC |
| 15:14.19 | jbevren | I have the old one |
| 15:14.29 | jbevren | but I want to get the 3e just to have more gates when I need em :) |
| 15:14.33 | davidc__ | yup, I've got the s3board.. and the s3eboard is on its way |
| 15:14.45 | jbevren | I got the hivolume kit (with the cpld kit) |
| 15:14.49 | davidc__ | also, the s3eboard has onboard ethernet |
| 15:14.52 | davidc__ | which I really like |
| 15:15.03 | davidc__ | just a phy wired to the fpga - which is what I want |
| 15:15.14 | davidc__ | .. really want one of those V4's - with a PPC and the GEMacs |
| 15:15.56 | jbevren | yep |
| 15:16.13 | jbevren | since its just a phy hooked to the fpga, you can use the line for transmitting more than ethernet hehe |
| 15:16.32 | davidc__ | yup |
| 15:16.52 | davidc__ | jbevren: well.. you'd need to disable autonegotiation |
| 15:16.55 | davidc__ | and a few other bits |
| 15:16.57 | jbevren | I want to get a gigabit phy for running serial extensions for (slow) parallel buses |
| 15:17.33 | davidc__ | jbevren: wouldn't you be better off with some of those bus->LVDS pairs things from natsemi? |
| 15:17.46 | jbevren | sure, but I'm not big on a 50-pair cable :) |
| 15:17.47 | davidc__ | you just shove a clocked paralell bus into it - out comes LVDS |
| 15:17.57 | davidc__ | er - its a serdes :P |
| 15:18.04 | davidc__ | so it upclocks |
| 15:19.36 | davidc__ | aka - it does exactly what you want - in one IC |
| 15:19.58 | davidc__ | dunno how fast it goes |
| 15:20.35 | jbevren | heh |
| 15:20.47 | jbevren | does it handle busmastering on both ends? |
| 15:21.41 | davidc__ | jbevren: nah, it unidirectional too |
| 15:21.46 | davidc__ | you just give it clock + data lines |
| 15:21.54 | davidc__ | it shoves it down 1 or 2 lvds pairs |
| 15:22.02 | davidc__ | and out pops the same clock + data @ the other end |
| 15:23.47 | jbevren | so thats not very helpful :) |
| 15:24.32 | davidc__ | eh - but you gotta makes sure your paralell bus can withstand the latency if its bidirectional |
| 15:26.44 | jbevren | you sid it was unidirectional |
| 15:26.49 | jbevren | is it unidirectional or bidirectional? |
| 15:26.56 | jbevren | those two are mutually exclusive |
| 15:28.20 | davidc__ | I meant in general |
| 15:28.36 | davidc__ | the Serdes'es from natsemi are uni, AFAIK |
| 15:28.57 | jbevren | ahh |
| 15:29.12 | davidc__ | most bidir busses I can think of wouldn't play nice with having a multi bus cycle latency between the ends |
| 15:29.27 | jbevren | yeah, but the target bus is only clocked at 1mhz |
| 15:29.31 | davidc__ | ah :P |
| 15:29.32 | jbevren | so latency shouldnt be a problem |
| 15:29.35 | davidc__ | then you've got tons of time |
| 15:29.37 | jbevren | I just need bidirectionality |
| 15:29.40 | davidc__ | as long as your link is fairly short |
| 15:29.40 | jbevren | yeah |
| 15:29.42 | jbevren | loads of time :) |
| 15:29.50 | jbevren | I'm not looking at going 30 feet :D |
| 15:29.57 | jbevren | maybe 1 foot |
| 15:29.59 | jbevren | 3 max |
| 15:30.01 | davidc__ | ah, :P |
| 15:30.10 | davidc__ | I thought you were talking 500, 600 feet ;) |
| 15:30.21 | jbevren | basically, I want to stick my old computer's accelerator in its own box so its not hanging off the back of the computer |
| 15:30.30 | davidc__ | accelerator? |
| 15:30.34 | jbevren | nod |
| 15:30.37 | davidc__ | as in graphics accelerator? |
| 15:30.39 | jbevren | internal cpu: 1mhz, accelerator: 20mhz |
| 15:30.41 | jbevren | <PROTECTED> |
| 15:30.47 | jbevren | it runs 21x faster |
| 15:31.16 | davidc__ | FPGA - capable of 200mhz - you could probably implement both the accelerator + the comp all inside the FPGA :P |
| 15:31.34 | jbevren | quite possible |
| 15:31.55 | jbevren | the accelerator is a 65816 cpu running at 20mhz, and a couple altera chips (one for bus management, the other for managing a stick of FPM ram) |
| 15:32.16 | sunrunner20 | thers a few issues w/ the TEMAC atm |
| 15:32.28 | sunrunner20 | just fyi |
| 15:32.39 | davidc__ | GEMAC? |
| 15:32.42 | jbevren | the trick is getting the timing exact, cause the disk software relies on cycle-exact timing with the floppy drive's cpu on a bitbanged 2-bit-wide serial bus |
| 15:32.45 | sunrunner20 | no |
| 15:32.46 | sunrunner20 | TEMAC |
| 15:32.49 | davidc__ | sunrunner20: TEMAC? |
| 15:32.51 | sunrunner20 | tri |
| 15:32.55 | sunrunner20 | 10/100/1000 |
| 15:33.01 | davidc__ | ah, Tri Ethernet Mac. Fine :P |
| 15:33.03 | jbevren | during disk IO, the accelerator changes its clock source to the computer's bus clock |
| 15:33.26 | sunrunner20 | quote "it works fine untill you add stuff to the PLB" |
| 15:34.15 | davidc__ | sunrunner20: joy. Note to self - don't use that part |
| 15:34.25 | sunrunner20 | lol |
| 15:34.32 | sunrunner20 | i think its soft issues |
| 15:34.52 | jbevren | davidc__: fyi, the device in question is a 'supercpu', an accelerator for commodore 64's |
| 15:34.53 | sunrunner20 | xilinx will have it fixed in a few weeks at most i think |
| 15:35.03 | jbevren | I use mine to run a unixoid multitasking OS called WiNGS |
| 15:35.27 | jbevren | that said, I want to dump the commodore hardware and rewrite the kernel to work with something more.. optimal. |
| 15:36.18 | jbevren | and to paraphrase the kernel author, there's very little that assumes there's only one cpu.. I'd love to try a smp setup hehe |
| 15:36.37 | sunrunner20 | damn you windows |
| 15:36.50 | davidc__ | sunrunner20: I really wish they'd make a V4FX without the rocketIO |
| 15:36.54 | davidc__ | thats what makes the price insane |
| 15:37.02 | davidc__ | well.. basically I want one in sub 100 bucks :P |
| 15:37.03 | sunrunner20 | its pageing like crazy and i have 400meg phyacial ram left |
| 15:37.07 | sunrunner20 | lol |
| 15:37.20 | jbevren | get more ram ) |
| 15:37.24 | sunrunner20 | lol |
| 15:37.27 | jbevren | I'll trade a V4 board for some ram |
| 15:37.29 | jbevren | :D |
| 15:37.33 | sunrunner20 | a gig is enough |
| 15:37.38 | jbevren | apparently |
| 15:37.39 | jbevren | a gig isnt enough |
| 15:37.43 | sunrunner20 | no |
| 15:37.43 | jbevren | youre swapping away :) |
| 15:37.46 | sunrunner20 | windows is jusg gay |
| 15:37.54 | jbevren | trust me, I know |
| 15:38.03 | jbevren | my laptop has 2gig and it pages with 1500M left |
| 15:38.47 | davidc__ | windows always pages |
| 15:38.54 | davidc__ | its insanely irritating |
| 15:39.24 | davidc__ | not to mention ISE -from xilinx [which leaks memory like a fucking pig] somehow manages to axe the OS enough that I need to reboot every few hours |
| 15:39.41 | davidc__ | btw sunrunner20 - tell your dad to tell Xilinx to make ISE stop sucking like a whore |
| 15:39.41 | jbevren | try running ISE on win2k in 128M ram |
| 15:39.48 | sunrunner20 | lol |
| 15:39.53 | davidc__ | hahah - it chugs on XP /w 1 Gig of ram |
| 15:40.02 | jbevren | davidc__: xp chugs on 1G of ram |
| 15:40.04 | jbevren | j/k :) |
| 15:40.24 | jbevren | I'd reinstall ise on my laptop but I dont have a usb jtagger |
| 15:40.28 | sunrunner20 | are y a'll closing all of ISE's pointless windows? |
| 15:40.40 | jbevren | I close any subwindow I'm not using |
| 15:40.50 | jbevren | normally just the file nav, project nav, log, and editor ar eopen |
| 15:40.51 | davidc__ | sunrunner20: yup |
| 15:40.59 | davidc__ | it just keeps allocating memory |
| 15:41.07 | jbevren | I -wanted- to use linux |
| 15:41.09 | davidc__ | + map + PnR sometimes take multi hundres of megs |
| 15:41.15 | jbevren | but the ise installer kept failing while trying to write to the cdrom |
| 15:41.20 | sunrunner20 | lol |
| 15:41.26 | davidc__ | sometimes PN itself sucks up almost a gig |
| 15:41.33 | davidc__ | and I've checked - its mostly just sitting in swap |
| 15:41.35 | sunrunner20 | hey |
| 15:41.43 | sunrunner20 | its still better than altera's stuff |
| 15:41.45 | davidc__ | never getting used - so there are some hugeass memory leaks in there |
| 15:41.57 | jbevren | I dunno, altera's stuff wasnt bad years ago wheN i played with it |
| 15:42.03 | davidc__ | ya, I concur |
| 15:42.04 | jbevren | man my shift timing is way out of whack |
| 15:42.23 | jbevren | the thing that's killing me is my ubicom SX dev kit |
| 15:42.27 | jbevren | I cant get it installed right |
| 15:42.37 | davidc__ | jbevren: hahaha - I've had lots of fun with that processor ;) |
| 15:42.51 | jbevren | davidc__: I need to build some ps2->parallel adapters for a friend |
| 15:43.00 | jbevren | he has an old vector graphics multiuser system for cp/m |
| 15:43.02 | davidc__ | got a 50k install base for some code I wrote for that |
| 15:43.18 | jbevren | it uses terminals that are basically composite monitors and parallel keyboards |
| 15:43.39 | davidc__ | sunrunner20: seriously - ISE might cost xilinx a design in - its memory leaks are just to crippling |
| 15:43.51 | davidc__ | sunrunner20: you might want to mention that to them |
| 15:44.00 | jbevren | He's got an old rs232 terminal that has bad ram in the raster engine, so I'm going to modify it to accept external video data ;-) |
| 15:44.00 | sunrunner20 | k |
| 15:44.10 | jbevren | and plug a sx20 in for the keyboard |
| 15:44.41 | davidc__ | sunrunner20: I'm doing a startup - it has the potential to go huge [well, everyone says they do] and the FPGA line we design in initially will probably be in it for a while |
| 15:45.11 | davidc__ | jbevren: btw - I've done 6Mbps throughput on a SX52 - /w 3 ethernet ports :P |
| 15:45.24 | davidc__ | jbevren: it sounds insane- but its doable |
| 15:46.13 | jbevren | well |
| 15:46.21 | jbevren | if you use busmastering ethernet chips that support fast switchign :) |
| 15:46.30 | jbevren | you never see the packet data |
| 15:46.38 | davidc__ | jbevren: haha.. if only ;) |
| 15:46.47 | jbevren | just look at the headers and tell chip A to throw the data into chip C's buffer |
| 15:46.49 | davidc__ | jbevren: I got to use stupid ISA RTL8019As'es |
| 15:46.51 | jbevren | tulips support that in linux :) |
| 15:47.12 | davidc__ | jbevren: my solution was to cheap and activate two devices on the ISA bus at once |
| 15:47.17 | davidc__ | strobe the read on one |
| 15:47.20 | jbevren | nod |
| 15:47.20 | davidc__ | then the write on another |
| 15:47.26 | davidc__ | so I never had to get it into mem |
| 15:47.27 | davidc__ | just onto the bus |
| 15:47.37 | jbevren | hehe |
| 15:47.54 | jbevren | I'd thought of such a thing before but never heard of it actually being used |
| 15:48.00 | jbevren | that'd make a handy ethernet repeater ;-) |
| 15:48.03 | davidc__ | I actually could do 10 - but some of the RTL chips failed @ that clocking |
| 15:48.16 | davidc__ | ya.. this was just a dumbass standalone firewall |
| 15:48.29 | jbevren | simple nat or transparent? |
| 15:48.34 | davidc__ | transparent |
| 15:48.37 | jbevren | cool |
| 15:48.38 | davidc__ | really really transparent |
| 15:48.40 | davidc__ | nah, not really |
| 15:48.42 | davidc__ | <PROTECTED> |
| 15:48.45 | jbevren | why not? :) |
| 15:48.54 | davidc__ | jbevren: eh, cool from a coding standpoint |
| 15:49.01 | jbevren | I built trans firewalls at a datacenter I used to work |
| 15:49.12 | jbevren | brb |
| 15:49.12 | davidc__ | but the company I did it for [and still work for, but not for long!] markets it whack |
| 15:49.14 | jbevren | nature break :) |
| 15:49.17 | davidc__ | "100% unhackable" |
| 15:49.38 | davidc__ | anyhow - I'm off to work.. back from there w/ davidc__[2] |
| 15:49.39 | jbevren | biggets challenge would be to not step on ftp etc hehe |
| 16:24.18 | *** join/#edev davidc__[2] (n=chatzill@s142-179-110-30.bc.hsia.telus.net) |
| 16:30.53 | *** join/#edev prpplague (n=billybob@72.22.137.214) |
| 16:31.13 | jbevren | wb billybob! |
| 16:31.14 | jbevren | hehe |
| 16:34.02 | davidc__[2] | *cough*redneck*ccough* |
| 16:37.11 | jbevren | hehe |
| 16:37.22 | jbevren | I'd say wb dave but I know of three daves here already |
| 16:37.34 | jbevren | bah :) |
| 17:26.15 | prpplague | hehe |
| 17:31.58 | *** join/#edev redblue (n=star@ppp161.108-253-207.mtl.mt.videotron.ca) |
| 18:13.47 | *** join/#edev Chocobo (n=swinchen@thecheat.eece.maine.edu) |
| 18:13.55 | Chocobo | Anyone here use ngspice for anything"? |
| 18:15.59 | davidc__[2] | Chocobo: why ask in two channels |
| 18:16.19 | davidc__[2] | Chocobo: when you haven't even waited for a response in the first? |
| 18:16.52 | Chocobo | Because someone might know in one channel and not the other. Less waiting, and I can listen to more than one person at a time. |
| 18:17.10 | jbevren | maybe its like applying for a job |
| 18:17.16 | jbevren | you ask several people at once |
| 18:17.28 | jbevren | cause its better to say no to anopportunity than miss some |
| 18:17.28 | Chocobo | exactly. |
| 18:17.29 | jbevren | shrug :) |
| 18:17.44 | Chocobo | well, in this case I would take both jobs. |
| 18:17.46 | Chocobo | lol |
| 18:17.51 | davidc__[2] | meh... I think of it as spamming.. but I guess not |
| 18:17.51 | jbevren | I wouldnt |
| 18:18.09 | jbevren | hehe |
| 18:18.13 | jbevren | unless theyre on alternating days |
| 18:24.27 | Chocobo | Here is the spice file I am having troubles with if anyone is interested. http://www.beneaped.org/hotwire.ckt |
| 18:28.12 | jbevren | http://www.ohgizmo.com/2006/04/28/new-line-of-bamboo-monitors-hits-the-market |
| 18:28.14 | jbevren | (/spam) |
| 19:57.45 | jbevren | zzz |
| 20:02.02 | agaffney | hurray! my 1U fans finally came today |
| 20:02.13 | agaffney | hopefully they'll be quite enough to let me run that switch 24/7 |
| 20:02.33 | jbevren | you didnt check their dbA rating before buying them? :) |
| 20:02.44 | agaffney | 22 |
| 20:02.49 | agaffney | but I really don't know what that means :) |
| 20:02.50 | jbevren | thats pretty quiet |
| 20:02.52 | agaffney | ok |
| 20:03.25 | agaffney | I think a tornado tearing through my apartment would be quiter than the current fans |
| 20:03.36 | agaffney | s/quiter/quieter/ |
| 20:03.47 | agaffney | purl: die |
| 20:03.49 | purl | ACTION takes two shots to the head and crumples to the ground, lifeless. |
| 20:18.08 | agaffney | ok, wtf |
| 20:18.11 | agaffney | I'm so confused |
| 20:18.16 | agaffney | these 40mm fans are way too small |
| 20:18.25 | agaffney | it looks like this baystack switch isn't 1U |
| 20:18.29 | agaffney | it's 2.75" height |
| 20:19.01 | agaffney | so these fans are way too small |
| 20:31.05 | *** join/#edev sorphin (n=dans@mewp.captainrock.com) |
| 20:31.05 | *** mode/#edev [+o sorphin] by ChanServ |
| 20:31.40 | agaffney | heh |
| 20:31.43 | agaffney | I've got plenty |
| 20:31.46 | agaffney | but I don't have a switch :( |
| 20:31.50 | agaffney | I got my fans in |
| 20:31.59 | agaffney | but it turns out this baystack is 2.75" height |
| 20:32.04 | agaffney | not 1U or 2U |
| 20:32.10 | agaffney | so these 1U fans obviously won't fit |
| 20:32.26 | agaffney | I just called EPC |
| 20:32.39 | sorphin | ah |
| 20:32.40 | sorphin | yeah |
| 20:32.45 | agaffney | the only used rackmountable switches they have are 10Mbps |
| 20:32.48 | sorphin | switch switch or kvm ? |
| 20:32.51 | sorphin | ah |
| 20:32.52 | agaffney | network switch |
| 20:33.09 | sorphin | well, rackmount is harder to get |
| 20:33.20 | sorphin | that's why my kvm was a pain |
| 20:33.25 | agaffney | I guess it doesn't *need* to be rackmount, but it would be more convenient |
| 20:33.25 | sorphin | but it's gonna have to go |
| 20:33.28 | sorphin | it's too small |
| 20:33.31 | sorphin | 4 port |
| 20:33.49 | agaffney | I'd get a KVM if I had more than 2 boxes that used PS/2+VGA :) |
| 20:34.04 | sorphin | heh |
| 20:34.16 | sorphin | well |
| 20:34.18 | sorphin | the mac mini |
| 20:34.22 | sorphin | the doze machine |
| 20:34.29 | sorphin | one for bringing up boxes |
| 20:34.39 | sorphin | and one for my main linux box |
| 20:34.42 | agaffney | the majority of my boxes are serial and/or 13w3 :) |
| 20:35.08 | agaffney | or VGA and USB |
| 20:38.08 | agaffney | hmm, I think I'll call EPC again and ask about any switches |
| 20:38.12 | agaffney | rackmountable or not |
| 20:38.31 | agaffney | I can just lay it on top of the netra if I really have to :) |
| 20:39.09 | sorphin | yeah |
| 20:39.33 | agaffney | we've got some decent netgear 24-port fanless switches here at work |
| 20:39.40 | agaffney | too bad they're ~$100 |
| 20:46.38 | agaffney | I had a better idea |
| 20:47.32 | agaffney | I posted a message to SLUUG's (St. Louis Unix Users' Group, if you didn't know) mailing list saying I was looking to trade any/all of the 3 baystack switches for a serial multiplexer and/or a 8/16-port network switch |
| 21:06.40 | sorphin | heh |
| 21:06.45 | sorphin | they don't come in 8/16 |
| 21:06.50 | sorphin | usually 12/24 ;) |
| 21:08.10 | agaffney | I've seen 8, 16, and 24 |
| 21:09.56 | sorphin | odd |
| 21:10.00 | sorphin | must be junk switches ;) |
| 21:10.10 | agaffney | heh |
| 21:30.42 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) |
| 21:33.24 | davidc__[2] | grah.. I just solved the most godawful embedded bug >:( |
| 21:35.56 | *** join/#edev prpplague (n=dave@72.22.141.65) |
| 21:43.53 | *** join/#edev Euler (n=swinchen@thecheat.eece.maine.edu) |
| 22:48.15 | *** join/#edev joe_bleau (n=joe_blea@adsl-69-152-192-31.dsl.fyvlar.swbell.net) |
| 22:50.23 | *** join/#edev ohsix (i=ohsix@bc111222.bendcable.com) |
| 22:55.59 | *** join/#edev redblue (n=star@ppp184.111-253-207.mtl.mt.videotron.ca) |