...saturday was all about the long awaited musical festivity unleashed upon the body corporate. after the unnecessary chaos of the bus transfer arrangement to the venue, the entire affair was, to put it mildly, good.
naturally, there were cringe-worthy moments, such as the second of the two special numbers that started the show, but that bears further discussion at a later time...
being somewhat delayed in my self-imposed deadline to fold all the g5 macintoshes into the glories of render slavery, i decided to go to work on sunday. figured that with the editorial crew (who use the machines) having their well-deserved sunday off (seeing as they did production assistance stuff in the run up to the show), i would have a better shot at finally configuring the g5s for their expanded role as render machines. so, checklist in hand (mental and otherwise), arrived at work about 3-ish in the afternoon.
an officemate decided to join me at work and set about investigating the animation complexities of an inverse-kinematic-based hair setup rig.
leaving him to it, i trudged away at the mac configurations, following all the steps, making sure that required directories were made, and various sundry settings input and verified.
it might just be me, but the mac interface and that benighted single button mouse are doing me no favors insofar as getting the work done quickly. granted i may just be more used to the three button logitech wheel mice that 'infest' our side of the floor, but for the life of me, i can't imagine how anyone would live through ctrl-alt-flower-single-click when a simple right mouse click would suffice (and quicker, too) for options.
so: three servers mounted to the root filesystem of the macintosh primary hard disk (analogous to the mapping methods employed on the linux render boxes) -- all in an attempt to more or less standardize the directory relationships across three different operating systems so that our render mongrel system can get its job done.
...and those three servers are not homogenous either. two intel boxes on red hat 9 linux, and one all-conquering dual g5 xserve running the server version of os x. now, that last part is a mouthful, but is actually a quite accurate way of describing that hardware/software combo.
which brings me back to the permission conundrum that had us backtracking off the original g4 xserve and hastily back to the old linux server. everything had been tried, permissions relaxed everywhere, and still our render system would not work.
this time around, we set about making sure the win2k boxes would work well with the xserve. then we moved on the render management system -- and that seemed to have been resolved by deactivating a key security component of os x server, something called 'light directory access protocol'. the troubleshooter noted that 'ldap' was looking to be the future of server permission management, but at this point in time with a deadline looming large, there was no point in trying to figure it out. so he disabled it. lo and behold, the render management system actually began to proceed beyond login failures.
it moved on to render failures.
wonderful.
and so it came to pass that i was logged in remotely on one of the linux boxes and i finally noticed the render system log mentioning that the mounted filesystems from the xserve were read only.
after some web research, found a way to get some info out of the xserve as far as how it was handling nfs (network file system) shares -- the primary method by which we access the servers from the linux/mac boxes. and yes, indeed, there was a switch in the options that was wonderful in its simplicity. the line read:
opts='ro'
hmm. web research showed that by default, it ought to be 'rw'.
ah, except that the example given was set to 'ro' with the note that by default, it ought to be 'rw'.
perhaps the troubleshooter in his haste had overlooked the note.
well, pointed this out to him saturday morning, and he went to the NOC forthwith and removed the offending option.
...but the time had arrived for the exodus to the musical extravaganza venue.
back to the tale at hand: g5s configured, i launched duplicate renders on all the farm and all the macs, and to make matters more colorful, threw in a 65GB copy from a linux server to the xserve where all the rendering machines were reading and writing to.
thence home, after dinner at italianni's. at some point in the dinner with the officemate (who by this time had resigned the attempt to come to terms with the hair rig) -- a frog keychain was placed on the table, together with a little laminated calling card.
the suspect swiftly moved elsewhere, the expected spiel not forthcoming.
i looked at the card, and it read (according to my dim memory)
i am deaf
...
...
...
if you are noble, you will purchase this ... for a hundred pesos
hm. so i am now the proud owner of a frog flashlight that bestows nobility.
---
and today, monday, i finally have some performance figures for the troubleshooter and the higher-ups in technical support. compared to a single dual-proc g5 rendering to a linux server, the xserve improves render performance by at least 10%. that's good. however, it appears to slow down the performance of the linux render farm, however slightly (about 5 minutes in 3 hours).
have to look at that closely.
so no sunday, no holiday; at least i have the frog of nobility.
Monday, November 29, 2004
Friday, November 19, 2004
shape of the beast
...well, not quite. ...yet.
to wit: 13 sequences. 13. yum. anyway, out of 13 sequences, 4 were on the fedex-ed hard disk. there was nothing else accompanying the hard disk, though there was a bunch of other stuff that we were expecting.
- 0 -
backtrack: today, we discovered that osx server has by default very rigorous security settings. settings that prevent our render management system from writing its important log files (useful for performance monitoring) -- hence, stopping any possibilities of rendering with the new server cold.
- 0 -
yesterday (thursday), it was finally cleared up why the linux render boxes could not mount the new server's shared resources -- instead of server:/storage/data, that osx server's path for the raid was server:/Volumes/data. and this was not known to me when the server was first activated for use the day before. anyway, that wrinkle was resolved by editing the appropriate entries in the linux render boxes' fstab files ("filesystem table", i'd guess).
at this time, with the new server's shared resources online, the question became: where's the old server? luckily, the troubleshooter extraordinaire was in the NOC when i decided to look. short discussion, and the agreement was that since the primary server was named after the capital planet of the star wars' old republic, it would make some sort of sense (though this was not, strictly speaking, really necessary) to name the old server, however temporarily, by another planet name in the star wars pantheon. hmm. nothing else came to mind but the planet that was blowed up good by the first death star in the first star wars film.
the rest of the day was spent (up to 8:30pm, actually -- long since going-home-time) tranferring, oh, maybe, 238GB of files from the old linux server box to the new osx server box. to its credit, the osx box does seem faster than the linux box -- but transferring is a two-way street. even if the osx box can handily run rings around the linux box in terms of hard disk performance and system throughput, the linux box itself is the limiting factor in how fast it can send data over the network.
- 0 -
day before yesterday (wednesday): prez walks in bright and early and inquires about new server. i have no clue. this is not received well. duh. i do some network trolling, and there it is: new server, old name; new workgroup. i wish they would inform me of these things after they had made the decisions so i'm not totally in the dark when the prez decides to ask me how things are going.
so, new server up, none of the proper shares. call up future prez, who says call up troubleshooter; so i bug the PM. after a long while, troubleshooter comes around, discuss, discuss; and share names agreed on. it doesn't take long before server is up with proper resource sharing.
all right! check the linux boxes to see if they mounted...
...did not. by this time, the magic hour has rolled around. so, next day, then.
- 0 -
tuesday.
the shape of the beast... ...is a standard 3.5 inch hdd. in that nice bubble wrap that some people like to pop. wonder how much of the project is on it? time will tell. considering that it took until the end of that day to finally hand it over to the troubleshooter for transfer to the as yet unknown new server.
- 0 -
back to the present.
not fun at all. first thing in the morning, attempt to submit a render through the render management software. there's an error i've never seen before. frenzied email later, troubleshooter arrives, early afternoon. we discuss the matter as best we can (i emailed the programmer of the software, and he had a deceptively simple suggestion) -- and we set out to try to attack the problem from as many angles as possible.
magic hour: no dice on my end. osx is weird enough for me, osx server is of another order of strangeness totally.
update 10:30pm. seems that the troubleshooter called it quits by 8:30 with no solutions in sight.
what's that standard quote from star wars? ah, yes. i have a bad feeling about this...
to wit: 13 sequences. 13. yum. anyway, out of 13 sequences, 4 were on the fedex-ed hard disk. there was nothing else accompanying the hard disk, though there was a bunch of other stuff that we were expecting.
- 0 -
backtrack: today, we discovered that osx server has by default very rigorous security settings. settings that prevent our render management system from writing its important log files (useful for performance monitoring) -- hence, stopping any possibilities of rendering with the new server cold.
- 0 -
yesterday (thursday), it was finally cleared up why the linux render boxes could not mount the new server's shared resources -- instead of server:/storage/data, that osx server's path for the raid was server:/Volumes/data. and this was not known to me when the server was first activated for use the day before. anyway, that wrinkle was resolved by editing the appropriate entries in the linux render boxes' fstab files ("filesystem table", i'd guess).
at this time, with the new server's shared resources online, the question became: where's the old server? luckily, the troubleshooter extraordinaire was in the NOC when i decided to look. short discussion, and the agreement was that since the primary server was named after the capital planet of the star wars' old republic, it would make some sort of sense (though this was not, strictly speaking, really necessary) to name the old server, however temporarily, by another planet name in the star wars pantheon. hmm. nothing else came to mind but the planet that was blowed up good by the first death star in the first star wars film.
the rest of the day was spent (up to 8:30pm, actually -- long since going-home-time) tranferring, oh, maybe, 238GB of files from the old linux server box to the new osx server box. to its credit, the osx box does seem faster than the linux box -- but transferring is a two-way street. even if the osx box can handily run rings around the linux box in terms of hard disk performance and system throughput, the linux box itself is the limiting factor in how fast it can send data over the network.
- 0 -
day before yesterday (wednesday): prez walks in bright and early and inquires about new server. i have no clue. this is not received well. duh. i do some network trolling, and there it is: new server, old name; new workgroup. i wish they would inform me of these things after they had made the decisions so i'm not totally in the dark when the prez decides to ask me how things are going.
so, new server up, none of the proper shares. call up future prez, who says call up troubleshooter; so i bug the PM. after a long while, troubleshooter comes around, discuss, discuss; and share names agreed on. it doesn't take long before server is up with proper resource sharing.
all right! check the linux boxes to see if they mounted...
...did not. by this time, the magic hour has rolled around. so, next day, then.
- 0 -
tuesday.
the shape of the beast... ...is a standard 3.5 inch hdd. in that nice bubble wrap that some people like to pop. wonder how much of the project is on it? time will tell. considering that it took until the end of that day to finally hand it over to the troubleshooter for transfer to the as yet unknown new server.
- 0 -
back to the present.
not fun at all. first thing in the morning, attempt to submit a render through the render management software. there's an error i've never seen before. frenzied email later, troubleshooter arrives, early afternoon. we discuss the matter as best we can (i emailed the programmer of the software, and he had a deceptively simple suggestion) -- and we set out to try to attack the problem from as many angles as possible.
magic hour: no dice on my end. osx is weird enough for me, osx server is of another order of strangeness totally.
update 10:30pm. seems that the troubleshooter called it quits by 8:30 with no solutions in sight.
what's that standard quote from star wars? ah, yes. i have a bad feeling about this...
Monday, November 15, 2004
tooth hurty revisited
so: 7 days later, 14 grams total of amoxicillin trihydrate and mefenamic acid (each), three pricks from an anesthetic-laden syringe needle, a surprisingly short period of -- wrenching? -- and the tooth was out.
interestingly, there was no pain to speak of, and i left the clinic still able to speak (not like the last visit i had at another dentist for some more, shall we say, advanced dentistry: i left her clinic with my entire mouth numb and immobile). getting home this time around was a cinch.
of course, now with a molar removed, i'll need something to take its place, and that's a cool 12K... ...ouch in the wallet.
...
the rest of the week ought to be interesting.
the data from the long awaited project ought to arrive -- so we'll finally behold the form of the beast that we have to vanquish...
...
interestingly, i met up with some old high-school barkada for lunch on sunday, and one of the odd thoughts that remains with me was a fragment of a conversation about some of the star wars novels
friend 1: "the writing can get so tiresomely predictable. for example, the author often uses the line 'with the horrible sounds of shrieking metal'..."
me: "well, that's better than 'with the pleasureable sounds of shrieking metal'..."
the others at the table were lucky not to have been drinking at the moment, given the laughter that greeted that comment.
interestingly, there was no pain to speak of, and i left the clinic still able to speak (not like the last visit i had at another dentist for some more, shall we say, advanced dentistry: i left her clinic with my entire mouth numb and immobile). getting home this time around was a cinch.
of course, now with a molar removed, i'll need something to take its place, and that's a cool 12K... ...ouch in the wallet.
...
the rest of the week ought to be interesting.
the data from the long awaited project ought to arrive -- so we'll finally behold the form of the beast that we have to vanquish...
...
interestingly, i met up with some old high-school barkada for lunch on sunday, and one of the odd thoughts that remains with me was a fragment of a conversation about some of the star wars novels
friend 1: "the writing can get so tiresomely predictable. for example, the author often uses the line 'with the horrible sounds of shrieking metal'..."
me: "well, that's better than 'with the pleasureable sounds of shrieking metal'..."
the others at the table were lucky not to have been drinking at the moment, given the laughter that greeted that comment.
Tuesday, November 09, 2004
tooth hurty / dvd afternoon
...well, whaddya know. couldn't get problem tooth extracted due to an infection in it -- so i'll have to take this antibiotic regimen for the next five days before going back and having the molar pulled.
course, i could have had it removed today, but the dentist couldn't guarantee that the anesthetic would hold, given the infection's presence. others had tried to have teeth extracted in a similar condition; but, and to quote: "sumuko sila."
not a pleasant thought.
so, got the prescription filled, and went home to get a head start on the treatment. also decided not to chase the half-day left if i were still to go to work; sleep may help the antibiotic do its voodoo.
ultimately, didn't sleep all that long -- never gotten used to sleeping in the afternoon. what then, to do.
hmm. shrek 2. ok, let's see what they've come up with.
...there are improvements over the first installment, true. a little better "acting," the effects were also good. there were moments that could elicit a smile, but overall, it just didn't somehow "ring true." in sum, i was unaffected by it. which, come to think of it, is similar to the effect don bluth's movies have on me -- not that he had anything to do with the movie, unless i'm mistaken.
i dunno. something was missing. and the soundtrack didn't help either. perhaps dreamworks is trying to out-disney disney with formulaic productions (and song and dance numbers) -- but disney's track record with it's in-house productions is surely a sign that formula isn't the way to go. and the disney-pixar deal is mostly a distribution deal on disney's part, as far as i know -- so the movies under that banner don't indicate much by way of disney's capability to come up with compelling movies.
maybe it was the fact that despite the improvement in their technology, they didn't attempt something different. it felt more like a concerted effort to cash in on a previous best-selling attempt. come to think of it, it may well have been that.
so, in sum: forgettable movie.
another movie that may not offer a payback for time and money spent seeing it -- polar express. tom hanks' voice acting talent notwithstanding, the preview in the theater showed what seems to be a repeat of the disaster that was final fantasy: the spirits within. well, as far as the clunky effect that a pure mocap setup has on ersatz animation. heck, in the shots shown, the eyes were decidedly dead-looking -- and cross-eyed to boot.
ok, then. shrek 2 has better animation than polar express. that's probably the best i can say for that.
after shrek 2, decided to watch katsuhiro otomo's akira. still stunning after all these years. even if the edition i have doesn't have any english dub or subtitles. interestingly, the credits note the use of 3-D software. ...from wavefront technologies, to boot. yes, the wavefront of alias|wavefront fame. and that was back in 1988. amazing the progress of technology. ...and sometimes, the wierd purposes they're put to: final fantasy/polar express...
just because it can be done in 3-D doesn't mean that it should.
...
gotta watch the incredibles again.
course, i could have had it removed today, but the dentist couldn't guarantee that the anesthetic would hold, given the infection's presence. others had tried to have teeth extracted in a similar condition; but, and to quote: "sumuko sila."
not a pleasant thought.
so, got the prescription filled, and went home to get a head start on the treatment. also decided not to chase the half-day left if i were still to go to work; sleep may help the antibiotic do its voodoo.
ultimately, didn't sleep all that long -- never gotten used to sleeping in the afternoon. what then, to do.
hmm. shrek 2. ok, let's see what they've come up with.
...there are improvements over the first installment, true. a little better "acting," the effects were also good. there were moments that could elicit a smile, but overall, it just didn't somehow "ring true." in sum, i was unaffected by it. which, come to think of it, is similar to the effect don bluth's movies have on me -- not that he had anything to do with the movie, unless i'm mistaken.
i dunno. something was missing. and the soundtrack didn't help either. perhaps dreamworks is trying to out-disney disney with formulaic productions (and song and dance numbers) -- but disney's track record with it's in-house productions is surely a sign that formula isn't the way to go. and the disney-pixar deal is mostly a distribution deal on disney's part, as far as i know -- so the movies under that banner don't indicate much by way of disney's capability to come up with compelling movies.
maybe it was the fact that despite the improvement in their technology, they didn't attempt something different. it felt more like a concerted effort to cash in on a previous best-selling attempt. come to think of it, it may well have been that.
so, in sum: forgettable movie.
another movie that may not offer a payback for time and money spent seeing it -- polar express. tom hanks' voice acting talent notwithstanding, the preview in the theater showed what seems to be a repeat of the disaster that was final fantasy: the spirits within. well, as far as the clunky effect that a pure mocap setup has on ersatz animation. heck, in the shots shown, the eyes were decidedly dead-looking -- and cross-eyed to boot.
ok, then. shrek 2 has better animation than polar express. that's probably the best i can say for that.
after shrek 2, decided to watch katsuhiro otomo's akira. still stunning after all these years. even if the edition i have doesn't have any english dub or subtitles. interestingly, the credits note the use of 3-D software. ...from wavefront technologies, to boot. yes, the wavefront of alias|wavefront fame. and that was back in 1988. amazing the progress of technology. ...and sometimes, the wierd purposes they're put to: final fantasy/polar express...
just because it can be done in 3-D doesn't mean that it should.
...
gotta watch the incredibles again.
Monday, November 08, 2004
high tech / third world = disconnect
phone line at home died friday night. thanks to the glories of cellular communication was able to phone in to globelines/innove's hotline. ...but not before a few false starts with my cellphone indicating that the hotline numbers were invalid. forgot to include the "02" before the number itself, for dialing a landline...
anyways, temper barely in check as answerdroid tells me first of all to check the connections outside the house... ...to which i say: okay, i can look at the wires, but what in the world would i divine by doing so? phone line troubleshooting is definitely out of my area of expertise -- this part, an internal monologue.
all right, sir, we'll try to send a crew to check in 24 to 48 hours...
...well, that's some level of precision.
saturday afternoon rolls around; lo, and behold! no, the phone's still not working; but the lineman shows up. he attaches some sort of doohickey gadget to the little box at the end of the phone line. i can hear the warbling sound it makes in his earpiece as he activates it. so: the connection inside the house is good. he goes out, i follow and watch as he climbs up a pole on a corner opposite our block. i go back inside and watch some tv. in a while, he comes back, and upon inquiry, reveals that the connections in the box on the utility pole 'had loosened'.
???
anyway, phone's working now. but had no time to compose nor post -- but i had intended to -- it would have likely been about the amazing absurdity of that happenstance. it's sorta unimaginable -- how do you install a telephone line in a junction box such that it will loosen over time? granted, there was a magnitude 5-some earthquake in the early part of october, but shouldn't the line have loosened then? ...ah, the mysteries of telecommunications (or the hands that maintain it) in this day and age...
...but enough of that.
the incredibles.
fantastic movie. had the same effect on me as the lord of the rings trilogy (and extended versions): it was possible to miss the fact that this was a seriously long movie (for an animated feature). about as long, i think, as katsuhiro otomo's akira, actually. a feast for the eye, and what a story... ...some of the parts that might seem obligatory to others went over without my cringing. then again, this director is the same guy who came up with the iron giant, that seriously underappreciated (in the US, anyway) gem of very well integrated 3-D and traditional animation (and, yes, the story was good too).
i think i'll watch the incredibles again, next time to go over the details. ...what are the chances, i wonder, of there being an extended director's cut?
anyways, temper barely in check as answerdroid tells me first of all to check the connections outside the house... ...to which i say: okay, i can look at the wires, but what in the world would i divine by doing so? phone line troubleshooting is definitely out of my area of expertise -- this part, an internal monologue.
all right, sir, we'll try to send a crew to check in 24 to 48 hours...
...well, that's some level of precision.
saturday afternoon rolls around; lo, and behold! no, the phone's still not working; but the lineman shows up. he attaches some sort of doohickey gadget to the little box at the end of the phone line. i can hear the warbling sound it makes in his earpiece as he activates it. so: the connection inside the house is good. he goes out, i follow and watch as he climbs up a pole on a corner opposite our block. i go back inside and watch some tv. in a while, he comes back, and upon inquiry, reveals that the connections in the box on the utility pole 'had loosened'.
???
anyway, phone's working now. but had no time to compose nor post -- but i had intended to -- it would have likely been about the amazing absurdity of that happenstance. it's sorta unimaginable -- how do you install a telephone line in a junction box such that it will loosen over time? granted, there was a magnitude 5-some earthquake in the early part of october, but shouldn't the line have loosened then? ...ah, the mysteries of telecommunications (or the hands that maintain it) in this day and age...
...but enough of that.
the incredibles.
fantastic movie. had the same effect on me as the lord of the rings trilogy (and extended versions): it was possible to miss the fact that this was a seriously long movie (for an animated feature). about as long, i think, as katsuhiro otomo's akira, actually. a feast for the eye, and what a story... ...some of the parts that might seem obligatory to others went over without my cringing. then again, this director is the same guy who came up with the iron giant, that seriously underappreciated (in the US, anyway) gem of very well integrated 3-D and traditional animation (and, yes, the story was good too).
i think i'll watch the incredibles again, next time to go over the details. ...what are the chances, i wonder, of there being an extended director's cut?
Thursday, November 04, 2004
frustration
...seems to be the theme for today.
caught a bus at just about the perfect time to get one; a few minutes after 8. it was a fairly fast trip -- at least until we got to alabang. the driver decided to not to line up where the buses generally stop; instead he made for the intersection where the light was currently red. naturally, people got off slowly -- as i got within three people of the door, light turns green. now, other than that initial stop, there are no real places to let people off (where you won't get caught, anyway).
so the bus decided to finally pause long enough for the rest of the passengers to get off (damnably, at a 'no unloading' zone just shy of the re-entry ramp to the south superhighway. i wanted to cuss and kick the driver's head in.
it was a full 12 minute walk back to the jeepney terminal, crossing at designated places, etc. another irritant was the small triangle of land between a "y" intersection close to the terminal. in it, the direct route from the crossing at one side to the other was barricaded so that the flow of pedestrians absolutely had to walk through a greenhills-style shop maze.
abso-bloomin-lutely irritating.
got across, and totally unwilling to indulge in the aggravating jeepney system of transportation, decided to take a cab.
luckily, an empty one passed by just as i got to the loading/unloading zone.
get to work -- and discover that the massively high-resolution render i had submitted to the farm had not rendered at all.
the resolution was 7500 by 5700 pixels. i was wondering if it was too high; tried it on my machine, and it quit with a 'not enough memory' error. tried using a render subregion switch with the render farm submit script -- which came back with an error that the option was unrecognized.
wtf?
it was in the documentation, how the hell could it be an unrecognized option?
after lots of gritted teeth web searches and trying all manner of options, it turns out that the software has two ways of rendering -- the main software itself which in turn calls a standalone render, and the standalone renderer itself invoked directly. and the kicker: command line switches for either one are different enough to cause unwarranted agony, especially if one has no idea why the render is failing.
arrrrrrrrrrrrrrrrrrrrrrrrrrrgh.
---
more tests, an sms or two, and the issue finally came to light. apparently, a convenience feature in the software had again reared its ugly head (this had happened before). it was all right when the windows machines worked with the linux server's 'windows' shared resource; but when a job was submitted using the linux farm (which uses an altogether different method of 'mapping' network shared data), the software feature would hang the render process altogether.
so, after a whole day gritting my teeth, got the render to work; and surprisingly, it rendered with a different render engine altogether (the renderer has four options of render 'engines' -- the one that seemed to work this time was one that required a license per processor). have to look at that unexpected thing later.
in the end, also decided to render the whole picture in one go, no more subregion shenanigans.
...not a pretty picture, though. still much work to be done.
at least, now i know how to render it.
end of day.
update
interestingly, the level of frustration that had built up over the day did not diminish with the trip home. not even 'the fairly oddparents' or 'spongebob squarepants' managed to take the edge off how i felt. just on a lark, and this is fairly odd for me (i've somehow lost some urge/need to see movies unless they're something i really want to see), i put the dvd of 'atlantis: the lost empire' on. 'course i fast-forwarded through the painful first few minutes and settled in to watch once they reached atlantis itself. i think it's still a good movie, even if it didn't perform as disney wanted it to. i'd have to agree with a friend's assessment that disney's adherence to a formula and a certain seemingly prescribed time limit in the end does more harm than good. at the same time, it can be fairly hard to say what'll sell and what won't. i liked 'iron giant' too, and that didn't do well either...
...does that say anything about my taste in movies? (",)
anyway, i'm feeling much better now. losing myself in the latter parts of atlantis certainly did much to improve my mood.
caught a bus at just about the perfect time to get one; a few minutes after 8. it was a fairly fast trip -- at least until we got to alabang. the driver decided to not to line up where the buses generally stop; instead he made for the intersection where the light was currently red. naturally, people got off slowly -- as i got within three people of the door, light turns green. now, other than that initial stop, there are no real places to let people off (where you won't get caught, anyway).
so the bus decided to finally pause long enough for the rest of the passengers to get off (damnably, at a 'no unloading' zone just shy of the re-entry ramp to the south superhighway. i wanted to cuss and kick the driver's head in.
it was a full 12 minute walk back to the jeepney terminal, crossing at designated places, etc. another irritant was the small triangle of land between a "y" intersection close to the terminal. in it, the direct route from the crossing at one side to the other was barricaded so that the flow of pedestrians absolutely had to walk through a greenhills-style shop maze.
abso-bloomin-lutely irritating.
got across, and totally unwilling to indulge in the aggravating jeepney system of transportation, decided to take a cab.
luckily, an empty one passed by just as i got to the loading/unloading zone.
get to work -- and discover that the massively high-resolution render i had submitted to the farm had not rendered at all.
the resolution was 7500 by 5700 pixels. i was wondering if it was too high; tried it on my machine, and it quit with a 'not enough memory' error. tried using a render subregion switch with the render farm submit script -- which came back with an error that the option was unrecognized.
wtf?
it was in the documentation, how the hell could it be an unrecognized option?
after lots of gritted teeth web searches and trying all manner of options, it turns out that the software has two ways of rendering -- the main software itself which in turn calls a standalone render, and the standalone renderer itself invoked directly. and the kicker: command line switches for either one are different enough to cause unwarranted agony, especially if one has no idea why the render is failing.
arrrrrrrrrrrrrrrrrrrrrrrrrrrgh.
---
more tests, an sms or two, and the issue finally came to light. apparently, a convenience feature in the software had again reared its ugly head (this had happened before). it was all right when the windows machines worked with the linux server's 'windows' shared resource; but when a job was submitted using the linux farm (which uses an altogether different method of 'mapping' network shared data), the software feature would hang the render process altogether.
so, after a whole day gritting my teeth, got the render to work; and surprisingly, it rendered with a different render engine altogether (the renderer has four options of render 'engines' -- the one that seemed to work this time was one that required a license per processor). have to look at that unexpected thing later.
in the end, also decided to render the whole picture in one go, no more subregion shenanigans.
...not a pretty picture, though. still much work to be done.
at least, now i know how to render it.
end of day.
update
interestingly, the level of frustration that had built up over the day did not diminish with the trip home. not even 'the fairly oddparents' or 'spongebob squarepants' managed to take the edge off how i felt. just on a lark, and this is fairly odd for me (i've somehow lost some urge/need to see movies unless they're something i really want to see), i put the dvd of 'atlantis: the lost empire' on. 'course i fast-forwarded through the painful first few minutes and settled in to watch once they reached atlantis itself. i think it's still a good movie, even if it didn't perform as disney wanted it to. i'd have to agree with a friend's assessment that disney's adherence to a formula and a certain seemingly prescribed time limit in the end does more harm than good. at the same time, it can be fairly hard to say what'll sell and what won't. i liked 'iron giant' too, and that didn't do well either...
...does that say anything about my taste in movies? (",)
anyway, i'm feeling much better now. losing myself in the latter parts of atlantis certainly did much to improve my mood.
Wednesday, November 03, 2004
when it rains
...it pours
or something like that. it was a logical outcome; slept later than normal, woke later than desirable. read: late. naturally, it all goes on from there, a cascade of lateness. however, i find that, from time to time, it seems that once the critical threshold is breached (the point at which one can still barely make it on time, even if it takes a jaw-popping taxi ride to do so -- both from a monetary and hair-raising standpoint; and not that i have much hair to raise, being bald), the tension associated with trying to make it on time lessens somewhat.
today, for example, i was 41 minutes late.
the pain, of course, comes later -- when the paycheck rolls around. reminds me, gotta check what the peso/minute value of being late is.
oh yes, and it was raining when i woke up. if there's something i hate more than the fact that rain raises the inconvenience factor of getting to work immensely, it would have to be soaked shoes in an airconditioned environment. then again, it would be a royal bummer to be completely soaked and be in an airconditioned environment.
so, to forestall such an eventuality, i waited for the rain to stop. hence the 'way late' today.
get to work, and i'm back in some sort of limbo, now that the render system has been massaged to work on all our platforms. so now i'm googling for stuff that hopefully will speed up the mac...
...more later.
or something like that. it was a logical outcome; slept later than normal, woke later than desirable. read: late. naturally, it all goes on from there, a cascade of lateness. however, i find that, from time to time, it seems that once the critical threshold is breached (the point at which one can still barely make it on time, even if it takes a jaw-popping taxi ride to do so -- both from a monetary and hair-raising standpoint; and not that i have much hair to raise, being bald), the tension associated with trying to make it on time lessens somewhat.
today, for example, i was 41 minutes late.
the pain, of course, comes later -- when the paycheck rolls around. reminds me, gotta check what the peso/minute value of being late is.
oh yes, and it was raining when i woke up. if there's something i hate more than the fact that rain raises the inconvenience factor of getting to work immensely, it would have to be soaked shoes in an airconditioned environment. then again, it would be a royal bummer to be completely soaked and be in an airconditioned environment.
so, to forestall such an eventuality, i waited for the rain to stop. hence the 'way late' today.
get to work, and i'm back in some sort of limbo, now that the render system has been massaged to work on all our platforms. so now i'm googling for stuff that hopefully will speed up the mac...
...more later.
Tuesday, November 02, 2004
render nirvana...
...well, not precisely.
there was something about the script component of the render management system (done in perl) that seemed somewhat lacking. this would turn out to be the final link in the chain to getting the whole shebang up and running.
...but not before i almost completely lost my cool (such as it is). the prez of the company showed up and i gave him the results of the render test that i had started last saturday, which showed (again, as i had already forgotten by this point) that the mac dual proc 2GHz g5's were performing at about the same level as a single p4 3GHz. he asked why, and i began that there were differing architectures... of course the prez could not believe this and wanted me to go to apple's site where all the claims are made, etc...
...it was a little too much for the barely begun day. i blurted that the tests were photoshop and that highlighted integer performance whereas we needed floating point for 3-D rendering -- and that macs are not precisely known for floating point performance.
now, this discussion has been going on for a while now -- but all the other times i've been pretty calm about it; but this time my frustration may have shown -- and this time the argument seems to have sunk in.
i can understand the disbelief -- after all, the macs do cost quite a bit compared to a clone p4 even with the bells and whistles. and with the rather lackluster render results i was obtaining, that meant the mac could no longer be considered as a competitive product for future render farm expansion.
the only remaining possible alternative to the intel inside experience (read: costly) to consider, then, was amd's opteron. which, given my generally positive experience with its athlon brethren, could be a good thing. especially if the price/performance slope is similar to the athlon versus p4.
i will admit the possibility that maybe there are yet undiscovered optimizations for the g5 processor and os x combo -- but the fact that nothing is obvious -- and even the web is shy of pages on optimization (not hard disk optimization) -- is definitely not a plus factor. one doesn't need to fight the os to get the maximum performance a platform has to offer.
afterwards, i went googling and a plethora of sites came up in the results that showed that the g5 was superior to the pc in what seemed to be a whole trainload of synthetic benchmarks. i say synthetic because while they do exercise the machine in predictable ways, the data they work with maybe does not reflect a real-world environment. and what's more real world that a project for a paying client?
in my tests, using a scene from the client's earlier data shipment to us, the mac steadfastly remained at a comparable level to the pc workstations. however that may be, i think i'll still hunt around for ways to get the mac up to it's supposed speed. must be the geek in me. (",)
anyways, back to the render script component. the render script would fail, noting that the 3-D software was not at the location the system environment variable said it would be. ...naturally, neither os x or the software installer for the 3-D software had any documentation where this environment variable could be located, much less modified.
at essentially wits' end, i decided to email the programmer of the render management system with the issues that i'd encountered. i asked if there was a way to add a third clause to the os check routine (which, in a nutshell checked if not NT, then Linux) such that the check would now be if not NT, check if OS X; if not that either, then Linux.
now, this is something i could have done myself, had i the documentation -- but, as i mentioned in a paragraph back, i had none.
...amazingly, the programmer emailed back, with a little piece of perl code that i spliced into the render script. there was a little snag, though, as he actually gave two options. suffice to say the simpler option only worked on the mac, making the linux render inoperable. with code spliced in, one render script could now be called to manage renders across three different os platforms. is good.
so, then. now we can set up the entire floor (pc's and macs) to augment the linux render farm. render nirvana. sort of.
hoarding and the skyway patrol
good thing that the building's guard noticed, and even better, warned us that the right rear wheel of rani's car was underinflated. so we spent some time changing the tire. jose took the lead in the process, but at a certain point, the guard got involved and helped expedite the wheel change. that done, rani zoomed us all to the drop off point where ferdz, louie and jose disembarked for their mrt transfer. i accompanied rani to her shop and thence made my way to a nearby supermarket.
that was the skyway patrol part: just writing it down; makes this post marginally longer, is all. hehehe
the hoarding bit has to do with 3-in-1 coffee. now, i could safely be considered a coffee addict -- if i have a mantra, it would be "coffee is good". i used to get 12-pack boxes of maxwell house sachets, but i'd go through a box in about a day or two, so it behooves me to get the 24-pack plastics. rani calls it hoarding (since i buy three plastics at a time), but i prefer the term "stocking up"...
oddly, though, haven't been able to find purefoods' cooked salami at either the 7-eleven near where i live, or the supermarket near rani's coffee shop. just thought it worth noting. a processed food item shortage?
there was something about the script component of the render management system (done in perl) that seemed somewhat lacking. this would turn out to be the final link in the chain to getting the whole shebang up and running.
...but not before i almost completely lost my cool (such as it is). the prez of the company showed up and i gave him the results of the render test that i had started last saturday, which showed (again, as i had already forgotten by this point) that the mac dual proc 2GHz g5's were performing at about the same level as a single p4 3GHz. he asked why, and i began that there were differing architectures... of course the prez could not believe this and wanted me to go to apple's site where all the claims are made, etc...
...it was a little too much for the barely begun day. i blurted that the tests were photoshop and that highlighted integer performance whereas we needed floating point for 3-D rendering -- and that macs are not precisely known for floating point performance.
now, this discussion has been going on for a while now -- but all the other times i've been pretty calm about it; but this time my frustration may have shown -- and this time the argument seems to have sunk in.
i can understand the disbelief -- after all, the macs do cost quite a bit compared to a clone p4 even with the bells and whistles. and with the rather lackluster render results i was obtaining, that meant the mac could no longer be considered as a competitive product for future render farm expansion.
the only remaining possible alternative to the intel inside experience (read: costly) to consider, then, was amd's opteron. which, given my generally positive experience with its athlon brethren, could be a good thing. especially if the price/performance slope is similar to the athlon versus p4.
i will admit the possibility that maybe there are yet undiscovered optimizations for the g5 processor and os x combo -- but the fact that nothing is obvious -- and even the web is shy of pages on optimization (not hard disk optimization) -- is definitely not a plus factor. one doesn't need to fight the os to get the maximum performance a platform has to offer.
afterwards, i went googling and a plethora of sites came up in the results that showed that the g5 was superior to the pc in what seemed to be a whole trainload of synthetic benchmarks. i say synthetic because while they do exercise the machine in predictable ways, the data they work with maybe does not reflect a real-world environment. and what's more real world that a project for a paying client?
in my tests, using a scene from the client's earlier data shipment to us, the mac steadfastly remained at a comparable level to the pc workstations. however that may be, i think i'll still hunt around for ways to get the mac up to it's supposed speed. must be the geek in me. (",)
anyways, back to the render script component. the render script would fail, noting that the 3-D software was not at the location the system environment variable said it would be. ...naturally, neither os x or the software installer for the 3-D software had any documentation where this environment variable could be located, much less modified.
at essentially wits' end, i decided to email the programmer of the render management system with the issues that i'd encountered. i asked if there was a way to add a third clause to the os check routine (which, in a nutshell checked if not NT, then Linux) such that the check would now be if not NT, check if OS X; if not that either, then Linux.
now, this is something i could have done myself, had i the documentation -- but, as i mentioned in a paragraph back, i had none.
...amazingly, the programmer emailed back, with a little piece of perl code that i spliced into the render script. there was a little snag, though, as he actually gave two options. suffice to say the simpler option only worked on the mac, making the linux render inoperable. with code spliced in, one render script could now be called to manage renders across three different os platforms. is good.
so, then. now we can set up the entire floor (pc's and macs) to augment the linux render farm. render nirvana. sort of.
hoarding and the skyway patrol
good thing that the building's guard noticed, and even better, warned us that the right rear wheel of rani's car was underinflated. so we spent some time changing the tire. jose took the lead in the process, but at a certain point, the guard got involved and helped expedite the wheel change. that done, rani zoomed us all to the drop off point where ferdz, louie and jose disembarked for their mrt transfer. i accompanied rani to her shop and thence made my way to a nearby supermarket.
that was the skyway patrol part: just writing it down; makes this post marginally longer, is all. hehehe
the hoarding bit has to do with 3-in-1 coffee. now, i could safely be considered a coffee addict -- if i have a mantra, it would be "coffee is good". i used to get 12-pack boxes of maxwell house sachets, but i'd go through a box in about a day or two, so it behooves me to get the 24-pack plastics. rani calls it hoarding (since i buy three plastics at a time), but i prefer the term "stocking up"...
oddly, though, haven't been able to find purefoods' cooked salami at either the 7-eleven near where i live, or the supermarket near rani's coffee shop. just thought it worth noting. a processed food item shortage?
Thursday, October 28, 2004
the week in review
well, found out firsthand why shorter rather than longer posts would be the norm for a blog: it does take a bit of work (and a fair amount of editing) to post -- and of course, if you're busy with something else; well, that takes precedence.
this week has been a case in point. happened to make a suggestion in a meeting about the upcoming rendering project about theoretically using the dual-proc mac g5s as part of the render farm in their off hours. suggestion received warmly -- naturally, guess who gets to prove the theory's feasibility...
boy, that took a fair bit of work -- and we're not out of the woods yet. first order of the day was to get the render management software to work on the mac. the installation instructions didn't work as advertised, and there was something that prevented the mac from being seen by the license manager for the render system. next step was to get the mac to connect the server's shared resources so as to appear as a folder in the root of the mac's hard drive (in similar fashion as we used on the linux/xeon render boxes).
i was trying to do both at the same time -- in hindsight, perhaps not the best way to go about it, but i had a deadline to come up with an answer.
and there were just so many variables. the mac itself, the render management software, the render license server on win2k, the main fileserver on linux, and our main 3-D software.
imagine the possibilities where the communication between these disparate packages would, could, and did, fail.
several times i gave up and handed the machine over to others and walked away. our future prez took a shot, and gave up. then came the resident troubleshooter of our umbrella company. to his credit, the troubleshooter managed to get the mac to be seen by the license server -- it was a deeply hidden setting on the mac side. the filesystem mount he ran out of time for -- it was already six-thirty on friday night; and there are problems with staying over beyond closing time...
saturday morning, i attacked the problem on the linux server side. lo and behold, it worked. so: license server visible, render management system daemon up and running, filesystem mounted; then the 3-D package wouldn't run. reinstalled that, and ready to fly.
go back to my machine, run the submit script... ...nothing. i'm almost out of time. ok, worse comes to worst. decide to run the render manually, off the command line. render running, and i begin to get some rendering statistics off a test scene i've been using as a benchmark for various machines.
...the results later.
later that day, paul and his fiancee wed. i had new leather shoes, and they were killing my feet severely. so i was somewhat distracted during the wedding proceedings themselves, and walking around church was a challenge.
however, the reception made up for the discomfort -- the food was great, and i overate (actually woke up at three am the following morning in some pain -- happens when i indulge overmuch in food).
anyway, to paul and rowena, congratulations! hope you enjoy our gift of 'light'. (",)
thus was the week.
this week has been a case in point. happened to make a suggestion in a meeting about the upcoming rendering project about theoretically using the dual-proc mac g5s as part of the render farm in their off hours. suggestion received warmly -- naturally, guess who gets to prove the theory's feasibility...
boy, that took a fair bit of work -- and we're not out of the woods yet. first order of the day was to get the render management software to work on the mac. the installation instructions didn't work as advertised, and there was something that prevented the mac from being seen by the license manager for the render system. next step was to get the mac to connect the server's shared resources so as to appear as a folder in the root of the mac's hard drive (in similar fashion as we used on the linux/xeon render boxes).
i was trying to do both at the same time -- in hindsight, perhaps not the best way to go about it, but i had a deadline to come up with an answer.
and there were just so many variables. the mac itself, the render management software, the render license server on win2k, the main fileserver on linux, and our main 3-D software.
imagine the possibilities where the communication between these disparate packages would, could, and did, fail.
several times i gave up and handed the machine over to others and walked away. our future prez took a shot, and gave up. then came the resident troubleshooter of our umbrella company. to his credit, the troubleshooter managed to get the mac to be seen by the license server -- it was a deeply hidden setting on the mac side. the filesystem mount he ran out of time for -- it was already six-thirty on friday night; and there are problems with staying over beyond closing time...
saturday morning, i attacked the problem on the linux server side. lo and behold, it worked. so: license server visible, render management system daemon up and running, filesystem mounted; then the 3-D package wouldn't run. reinstalled that, and ready to fly.
go back to my machine, run the submit script... ...nothing. i'm almost out of time. ok, worse comes to worst. decide to run the render manually, off the command line. render running, and i begin to get some rendering statistics off a test scene i've been using as a benchmark for various machines.
...the results later.
later that day, paul and his fiancee wed. i had new leather shoes, and they were killing my feet severely. so i was somewhat distracted during the wedding proceedings themselves, and walking around church was a challenge.
however, the reception made up for the discomfort -- the food was great, and i overate (actually woke up at three am the following morning in some pain -- happens when i indulge overmuch in food).
anyway, to paul and rowena, congratulations! hope you enjoy our gift of 'light'. (",)
thus was the week.
Wednesday, October 27, 2004
the mind wanders / my very own titanic adventure...
...except for the part where the ship sinks, else i wouldn't be here to tell the tale.
jose's post yesterday in his blog about being in cebu got me thinking about island-hopping days of yore. other than luzon, the only islands i've ever been on in this archipelago are palawan and mindoro (for puerto galera's white beach -- an excellent place for comet watching).
i'd like to see the famed beach-y perfection that is supposed to be boracay; but the reputation also has it's own reputation to be quite dear on the wallet. so that's a pipe dream... ...at the moment.
anyway, my first trip to palawan was engendered by the encouragement of paul, a friend and officemate. his fiancee was working at the legend hotel in puerto princesa, palawan; and his trip had been in the offing for a while. i, on my part, took a while to decide. so it was very last minute that i ended up booking a berth on a negros navigation ship scheduled to leave the weekend before the holy week observations that year. paul had taken a flight -- i could've too, short of the indecision.
besides, it would be an experience. what could go wrong?
...nothing went wrong. it went weird.
before the weirdness: it would be a literal voyage of firsts for me. previous travels had been with family and/or friends. this time, i'd be on my own. also first time on an inter-island vessel larger than the scaled-up bancas that ply the batangas-puerto galera route. and longest trip, too. 12 hours. that's just an hour less than a non-stop mnl/lax flight -- at a huge speed disparity. i'm sure that the ship i took is a perfectly swift vessel in it's own right -- and the operator certainly wouldn't operate it at maximum velocity... ...of course, that would be an experience too.
got to the port early, boarded with little fuss. (this is pre 9/11, of course. don't know how it is these days. have never taken another similar ship trip again). am in a room with two double-deck bunks flanking the door, with a diner-style table under a fixed picture window; benches attached to the walls in line with the bunks. just before departure, a couple came in; talking to one of the crew. the crewman went away and came back with news that a suite was available. away the couple went...
...and that meant that the cabin was all mine. (",)
to the weirdness:
passing corregidor. hm. take glasses off, and use binoculars. ship-shaped islands. concrete. whatever for? door opens. turn to look, nearsightedness allows me to see blur in doorway. door closes. what was that about?
later, lying in a lower bunk, reading a book. door opens, person walks straight in and sits by the window. door remains open.
hmm. i close the door. cliches about strained silences apply. i don't remember who speaks first, but the story: he's on his way to puerto princesa to do something about a show his mom is putting on at a hotel. his companions are workers for his mom, and they've gone gallivanting about the ship, locking the cabin, and he didn't have a key. so, he asked, kinda belatedly, if he could sit here for a while.
...
oookaaay... ...anyway, be nice, chitchat, etc. he leaves after a while, saying he'd be back to
accompany me to dinner, and never comes back. which is good.
after dinner, head out on deck for some fresh air. i had watched the sunset earlier, and it was quite picturesque. the night was breathtaking. there were a lot of small, puffy clouds about, and the moon was full. the 'jesus rays' one commonly associates with the sun behind clouds were in abundance, but this time the moon was powering them. great splotches of moonlit water with wavelets twinkling in that light, connected to the sky by vast columns of moonbeams. and in the shadows, darker patches in the water -- not sure what they were, but the boat seemed to be navigating an obvious path of lighter shaded water among these. the water is also rather calm. there are no swells, so the boat just seems to hang in space while the scenery floats by. quite peaceful.
look in on the bridge, and it's dark. all of the light in there was coming from the intruments. everyone was standing; no one was saying anything, and about five of them were at the forward windows, peering into the night.
decide to call it a night. sleep was fitful -- kept waking up to a bright light in the window that would move erratically. thinking about it, it was probably a planet or a star...
breakfast. i was early, hoping for better pickings than the night before... food still not good. scrambled eggs are cold. ah well.
...and then he shows up. with nothing better to do and only a couple more hours before we get to the port, i decide to accompany him around the ship. we end up at the bridge -- lo and behold, there are rungs leading up to the roof. up he goes. what!?
...don't know why, but up i go too. next thing, he hands me his camera, and i take a picture of him doing the 'king of the world' bit at the front of the roof. next thing, a crewman pops up and tells us it's forbidden to be up here... ...picture's taken, so that's an accomplishment for leo.
...why leo? i don't remember when, but sometime during the trip, he confided that his girlfriends have suggested that he resembled the titanic star... ...i may post a picture sometime, and you get to decide.
trip over, boat docks. paul and his fiancee are at the port, and as i go to meet them, leo shows up.
!!!
paul has made plans already, and we're discussing these -- and leo pipes up to the effect that he'd like to hang out with us for a while; he'd ask his mom to excuse him from the show or something like that...
...and the saga continues.
next time.
jose's post yesterday in his blog about being in cebu got me thinking about island-hopping days of yore. other than luzon, the only islands i've ever been on in this archipelago are palawan and mindoro (for puerto galera's white beach -- an excellent place for comet watching).
i'd like to see the famed beach-y perfection that is supposed to be boracay; but the reputation also has it's own reputation to be quite dear on the wallet. so that's a pipe dream... ...at the moment.
anyway, my first trip to palawan was engendered by the encouragement of paul, a friend and officemate. his fiancee was working at the legend hotel in puerto princesa, palawan; and his trip had been in the offing for a while. i, on my part, took a while to decide. so it was very last minute that i ended up booking a berth on a negros navigation ship scheduled to leave the weekend before the holy week observations that year. paul had taken a flight -- i could've too, short of the indecision.
besides, it would be an experience. what could go wrong?
...nothing went wrong. it went weird.
before the weirdness: it would be a literal voyage of firsts for me. previous travels had been with family and/or friends. this time, i'd be on my own. also first time on an inter-island vessel larger than the scaled-up bancas that ply the batangas-puerto galera route. and longest trip, too. 12 hours. that's just an hour less than a non-stop mnl/lax flight -- at a huge speed disparity. i'm sure that the ship i took is a perfectly swift vessel in it's own right -- and the operator certainly wouldn't operate it at maximum velocity... ...of course, that would be an experience too.
got to the port early, boarded with little fuss. (this is pre 9/11, of course. don't know how it is these days. have never taken another similar ship trip again). am in a room with two double-deck bunks flanking the door, with a diner-style table under a fixed picture window; benches attached to the walls in line with the bunks. just before departure, a couple came in; talking to one of the crew. the crewman went away and came back with news that a suite was available. away the couple went...
...and that meant that the cabin was all mine. (",)
to the weirdness:
passing corregidor. hm. take glasses off, and use binoculars. ship-shaped islands. concrete. whatever for? door opens. turn to look, nearsightedness allows me to see blur in doorway. door closes. what was that about?
later, lying in a lower bunk, reading a book. door opens, person walks straight in and sits by the window. door remains open.
hmm. i close the door. cliches about strained silences apply. i don't remember who speaks first, but the story: he's on his way to puerto princesa to do something about a show his mom is putting on at a hotel. his companions are workers for his mom, and they've gone gallivanting about the ship, locking the cabin, and he didn't have a key. so, he asked, kinda belatedly, if he could sit here for a while.
...
oookaaay... ...anyway, be nice, chitchat, etc. he leaves after a while, saying he'd be back to
accompany me to dinner, and never comes back. which is good.
after dinner, head out on deck for some fresh air. i had watched the sunset earlier, and it was quite picturesque. the night was breathtaking. there were a lot of small, puffy clouds about, and the moon was full. the 'jesus rays' one commonly associates with the sun behind clouds were in abundance, but this time the moon was powering them. great splotches of moonlit water with wavelets twinkling in that light, connected to the sky by vast columns of moonbeams. and in the shadows, darker patches in the water -- not sure what they were, but the boat seemed to be navigating an obvious path of lighter shaded water among these. the water is also rather calm. there are no swells, so the boat just seems to hang in space while the scenery floats by. quite peaceful.
look in on the bridge, and it's dark. all of the light in there was coming from the intruments. everyone was standing; no one was saying anything, and about five of them were at the forward windows, peering into the night.
decide to call it a night. sleep was fitful -- kept waking up to a bright light in the window that would move erratically. thinking about it, it was probably a planet or a star...
breakfast. i was early, hoping for better pickings than the night before... food still not good. scrambled eggs are cold. ah well.
...and then he shows up. with nothing better to do and only a couple more hours before we get to the port, i decide to accompany him around the ship. we end up at the bridge -- lo and behold, there are rungs leading up to the roof. up he goes. what!?
...don't know why, but up i go too. next thing, he hands me his camera, and i take a picture of him doing the 'king of the world' bit at the front of the roof. next thing, a crewman pops up and tells us it's forbidden to be up here... ...picture's taken, so that's an accomplishment for leo.
...why leo? i don't remember when, but sometime during the trip, he confided that his girlfriends have suggested that he resembled the titanic star... ...i may post a picture sometime, and you get to decide.
trip over, boat docks. paul and his fiancee are at the port, and as i go to meet them, leo shows up.
!!!
paul has made plans already, and we're discussing these -- and leo pipes up to the effect that he'd like to hang out with us for a while; he'd ask his mom to excuse him from the show or something like that...
...and the saga continues.
next time.
Tuesday, October 26, 2004
hidden gems among the rugby
yes, that rugby -- the contact cement that's the reputed mood-enhancer of choice for the non-jet-set who're into recreational pharmaceuticals.
not that the inhalation was exactly by choice, mind. our pantry's vinyl wood-plank-look tiles suffered from, shall we say, a failure to bond with the concrete flooring due to some, ahem, septic seepage, in turn due to old, easily clogged pipes (or so we were told). fancy that -- refurbished building, old pipes...
the liquid event was barely a week into our occupancy of this floor -- and immediately after the tiles were laid in the first place. well, today, a week or so later, some brilliant type decided that now would be a good time to re-acquaint the tiles with the floor -- so in treads a worker and dutifully introduces the rubber cement into the tile/floor equation... ...in a room with completely recirculated air courtesy of the air conditioning.
anyway, can't say whether the side-effect to the contact cement's solvent has taken hold of the others -- all i have is yet another headache.
speaking of headaches, now may be a good time to give a code name to the mocap software, so there'll be a simpler reference to it. i dub thee 'reyna'. close enough to the real name -- and a terrible pun to boot. (",)
mocap, as i've encountered it, is about markers. these markers are placed in, well, places on the body of the mocap model. various technologies exist to take the markers' position in 2-D space (the multiple camera ccds) and turn it into movement in 3-D space -- and it gets saved into a file for me (and any other hapless mocap cleanup person) to load and analyze and so on and so forth.
now, by themselves, the markers are for all intents and purposes tracking their individual destinies in some virtual space within the mocap software. playing the file, you can just barely make out the humanoid arrangement of the marker constellations. here's where connecting markers -- much like plotting constellations in the sky -- makes divining movement easier.
last week, i had managed to make a little program in reynascript that fairly automated this constellation process. it was, though, quite straightforward and brute force.
where do the gems in rugby come in?
well, solvent-laced air notwithstanding, i cut some code from the un-commented, un-documented pre-built reynascripts and folded it into a more elegant thing that did the same thing as before -- with the exception that i didn't have to manually include the group marker names in the script itself.
now i'm wondering if there's a way to catch errors. turns out that the improved constellation script just ups and dies when (predictably) some predefined marker names are modified or lost.
still, a good thing.
hmm. maybe there is more to this rugby bit...
...i'll never know.
not that the inhalation was exactly by choice, mind. our pantry's vinyl wood-plank-look tiles suffered from, shall we say, a failure to bond with the concrete flooring due to some, ahem, septic seepage, in turn due to old, easily clogged pipes (or so we were told). fancy that -- refurbished building, old pipes...
the liquid event was barely a week into our occupancy of this floor -- and immediately after the tiles were laid in the first place. well, today, a week or so later, some brilliant type decided that now would be a good time to re-acquaint the tiles with the floor -- so in treads a worker and dutifully introduces the rubber cement into the tile/floor equation... ...in a room with completely recirculated air courtesy of the air conditioning.
anyway, can't say whether the side-effect to the contact cement's solvent has taken hold of the others -- all i have is yet another headache.
speaking of headaches, now may be a good time to give a code name to the mocap software, so there'll be a simpler reference to it. i dub thee 'reyna'. close enough to the real name -- and a terrible pun to boot. (",)
mocap, as i've encountered it, is about markers. these markers are placed in, well, places on the body of the mocap model. various technologies exist to take the markers' position in 2-D space (the multiple camera ccds) and turn it into movement in 3-D space -- and it gets saved into a file for me (and any other hapless mocap cleanup person) to load and analyze and so on and so forth.
now, by themselves, the markers are for all intents and purposes tracking their individual destinies in some virtual space within the mocap software. playing the file, you can just barely make out the humanoid arrangement of the marker constellations. here's where connecting markers -- much like plotting constellations in the sky -- makes divining movement easier.
last week, i had managed to make a little program in reynascript that fairly automated this constellation process. it was, though, quite straightforward and brute force.
where do the gems in rugby come in?
well, solvent-laced air notwithstanding, i cut some code from the un-commented, un-documented pre-built reynascripts and folded it into a more elegant thing that did the same thing as before -- with the exception that i didn't have to manually include the group marker names in the script itself.
now i'm wondering if there's a way to catch errors. turns out that the improved constellation script just ups and dies when (predictably) some predefined marker names are modified or lost.
still, a good thing.
hmm. maybe there is more to this rugby bit...
...i'll never know.
Monday, October 25, 2004
monday meltdown
the arc of the day was rather short. a discussion and two meetings in the morning -- and i was spent.
oh, yeah. woke up with a headache -- maybe that had to do with having some milk the evening before, to help me get to sleep. milk, after all, is rich in calcium -- and my hypertension medicine is technically a 'calcium channel blocker'. so, to my layman's mind, those are probably not good things to mix. milk, much as i like the stuff, is nowadays a double-barrel of problems. lactose intolerance and the hypertension medicine... ...ugh.
when it rains, it pours: the long awaited project that saw the group's assembly is now on the cusp of finally arriving. although not quite in the form expected. two hurricanes, a brain cancer(?) and some money matters have resulted in the latter part of the production reaching us. rendering, compositing and, unless i'm mistaken, the soundtrack. still, it's a ton of work. 44 minutes of rendering and composite -- in a month and a half. the company will have to seriously up the render farm units to make that deadline.
then we have the other stuff.
one of which is the motion capture cleanup software. an oddly named software at that. then again, they're all oddly named. our main package is named after an extinct civilization -- and its main competition (which i hold an irrational bias for) is a three letter non-acronym. our compositing package is one in a family of products that could safely be categorized as a full-blown conflagration... ...about the only thing that's honestly, if somewhat unimaginatively, named is the editing software. and that's called -- and i paraphrase here -- "the end" professional. hmm. maybe that hit the nail on the head. imaginatively (or not) named.
back to the mocap peeve. i've been wrapping my mind (however unwilling -- it's that 'old dog new tricks' saw) around it for a couple of weeks now. it's made all the more, shall we say, challenging, by the fact that the documentation/tutorials are almost a version and a half older than the evaluation software we have. talk about lost in translation. they changed some syntax, they deprecated some functions, and they left no documentation of either. so slogging through the tutorials strikes me as being somewhat akin to archaeology. you could also think of it this way: you've a recipe written in language version 1.7. but your food replicator is version 2.3 and is not 100% backward compatible. imagine the joyous food you'd get out of that combination.
ok, so that's not an accurate picture, but it'll suffice... hehehe
anyway, after much progress over the past week, somehow the day's meetings left me in no state to interpret the hieroglyphs of the mocap tutorials. not that i was really feeling well to begin with. by the afternoon, my hands were colder and wetter than normal, and i was feeling very hot...
...but not as heated as when i have to confront the works of, i'm told, an ex-contributing-editor of the local version of a british car mag. that, then, can be considered good news. leastwise, i can now peacefully buy the next issues without fear of encountering strangeness like 'over productive wedding vegetables' in reference to the volvo hybrid station wagon/minivan/suv. granted, that quoted line didn't make it to the magazine (yay!) -- it is, however on the website. read it and weep. (isn't it ironic how much advertising i'm giving the site?)
(",)
at any rate, the rest of the afternoon was a lost cause insofar as the mocap was concerned. made very little headway.
and thus ended the day. odd, no? i began the post noting a short arc of the day -- and look how long this got.
oh, yeah. woke up with a headache -- maybe that had to do with having some milk the evening before, to help me get to sleep. milk, after all, is rich in calcium -- and my hypertension medicine is technically a 'calcium channel blocker'. so, to my layman's mind, those are probably not good things to mix. milk, much as i like the stuff, is nowadays a double-barrel of problems. lactose intolerance and the hypertension medicine... ...ugh.
when it rains, it pours: the long awaited project that saw the group's assembly is now on the cusp of finally arriving. although not quite in the form expected. two hurricanes, a brain cancer(?) and some money matters have resulted in the latter part of the production reaching us. rendering, compositing and, unless i'm mistaken, the soundtrack. still, it's a ton of work. 44 minutes of rendering and composite -- in a month and a half. the company will have to seriously up the render farm units to make that deadline.
then we have the other stuff.
one of which is the motion capture cleanup software. an oddly named software at that. then again, they're all oddly named. our main package is named after an extinct civilization -- and its main competition (which i hold an irrational bias for) is a three letter non-acronym. our compositing package is one in a family of products that could safely be categorized as a full-blown conflagration... ...about the only thing that's honestly, if somewhat unimaginatively, named is the editing software. and that's called -- and i paraphrase here -- "the end" professional. hmm. maybe that hit the nail on the head. imaginatively (or not) named.
back to the mocap peeve. i've been wrapping my mind (however unwilling -- it's that 'old dog new tricks' saw) around it for a couple of weeks now. it's made all the more, shall we say, challenging, by the fact that the documentation/tutorials are almost a version and a half older than the evaluation software we have. talk about lost in translation. they changed some syntax, they deprecated some functions, and they left no documentation of either. so slogging through the tutorials strikes me as being somewhat akin to archaeology. you could also think of it this way: you've a recipe written in language version 1.7. but your food replicator is version 2.3 and is not 100% backward compatible. imagine the joyous food you'd get out of that combination.
ok, so that's not an accurate picture, but it'll suffice... hehehe
anyway, after much progress over the past week, somehow the day's meetings left me in no state to interpret the hieroglyphs of the mocap tutorials. not that i was really feeling well to begin with. by the afternoon, my hands were colder and wetter than normal, and i was feeling very hot...
...but not as heated as when i have to confront the works of, i'm told, an ex-contributing-editor of the local version of a british car mag. that, then, can be considered good news. leastwise, i can now peacefully buy the next issues without fear of encountering strangeness like 'over productive wedding vegetables' in reference to the volvo hybrid station wagon/minivan/suv. granted, that quoted line didn't make it to the magazine (yay!) -- it is, however on the website. read it and weep. (isn't it ironic how much advertising i'm giving the site?)
(",)
at any rate, the rest of the afternoon was a lost cause insofar as the mocap was concerned. made very little headway.
and thus ended the day. odd, no? i began the post noting a short arc of the day -- and look how long this got.
Sunday, October 24, 2004
there's always a first time...
...isn't there?
at any rate, the 'x' instead of 's' for singularity -- the latter was already taken, hence the somewhat similar sounding 'x' in place. well, if you believe the pronunciation posted on hyperdic.com for 'xavier' anyhow: zey'vyer -- unlike those idiots who dub x-men who pronounce it ekszey'vyer. but hey, it's a free world (and not that i watch the x-men cartoons on tv, just running across ads while channel surfing can get my goat).
besides, i happen to like x (think what you will. hehehe). x marks the spot, etc, etc. besides, in physics, what is a singularity except the ultimate full stop to the universe as we know it -- a point of infinite density or somesuch. odd, isn't it -- if current views of physics and cosmology are to be believed, the ultimate fate of the universe might likely be a cloud of infinitely dense points embedded in an ever-expanding emptiness, all in darkness; with naught but a distant memory of light in the form of microwaves (could be wrong on the microwave light-memory, though). makes for an interesting mind-picture, nevertheless (if you happen to go for those things, anyway).
so, why singularity (other than being my ostensible civil status)? well, i have this habit of concentrating on a given task almost to the exclusion of nearly anything else (does that make me single-track-minded?). so what will most likely happen in this blog is what could be long stretches of posts on the current item of study and/or pet peeve about current item of study. or something to that effect.
then there's the debate on the writing skill (or lack of it) of a certain ulysses ang of motioncars.com (god, what a name for an automotive website. imagine it.) whose 'work' is also published by the local version of the bbc magazine top gear... (much to the hair-pulling of his editors, no doubt). don't get me started... ...oh, but i already did. hehehe.
hmm. make sense? oh well.
so here we are, first faltering steps into the world of the blogosphere.
wish me luck.
at any rate, the 'x' instead of 's' for singularity -- the latter was already taken, hence the somewhat similar sounding 'x' in place. well, if you believe the pronunciation posted on hyperdic.com for 'xavier' anyhow: zey'vyer -- unlike those idiots who dub x-men who pronounce it ekszey'vyer. but hey, it's a free world (and not that i watch the x-men cartoons on tv, just running across ads while channel surfing can get my goat).
besides, i happen to like x (think what you will. hehehe). x marks the spot, etc, etc. besides, in physics, what is a singularity except the ultimate full stop to the universe as we know it -- a point of infinite density or somesuch. odd, isn't it -- if current views of physics and cosmology are to be believed, the ultimate fate of the universe might likely be a cloud of infinitely dense points embedded in an ever-expanding emptiness, all in darkness; with naught but a distant memory of light in the form of microwaves (could be wrong on the microwave light-memory, though). makes for an interesting mind-picture, nevertheless (if you happen to go for those things, anyway).
so, why singularity (other than being my ostensible civil status)? well, i have this habit of concentrating on a given task almost to the exclusion of nearly anything else (does that make me single-track-minded?). so what will most likely happen in this blog is what could be long stretches of posts on the current item of study and/or pet peeve about current item of study. or something to that effect.
then there's the debate on the writing skill (or lack of it) of a certain ulysses ang of motioncars.com (god, what a name for an automotive website. imagine it.) whose 'work' is also published by the local version of the bbc magazine top gear... (much to the hair-pulling of his editors, no doubt). don't get me started... ...oh, but i already did. hehehe.
hmm. make sense? oh well.
so here we are, first faltering steps into the world of the blogosphere.
wish me luck.
Subscribe to:
Posts (Atom)
