Jump to content

taks

Members
  • Posts

    1960
  • Joined

  • Last visited

Everything posted by taks

  1. E6600, P5B-E mobo, 4 GB OCZ RAM (2x2), an 8800 GTS 320 MB and a thermaltake 600 W supply on the way. sooner than later i suppose. my wife will get much of my current rig, including the new 7900 GS vid card, and her old rig will become the "standby." i think i'll use it for something... dunno what yet. taks
  2. i'm about to get updated myself, btw. i dread the thought, but i need to invest the money for a legitimate, non-gaming related purpose: the simulation i'm writing for my dissertation. as it stands, it takes 5-10 minutes to do a 100 iteration monte carlo simulation. at some point, we're going to test bit-error rates, which will require 50k iterations (512 "bits" per iteration) which works out to nearly a week per run. not feasible. there are some simplifications i'll write to speed things up, but also some complications i have to add such as channel modeling (search for the term "rayleigh fading" if you're interested). at any rate, the 3200+ AMD processor with 1 GB of memory ain't up to snuff. particularly if i want to do other things (such as research, or writing) while a simulation is running. currently, when matlab runs, it takes over. nothing else can happen. i can barely move the mouse. not so on the core2 here at the office, however. i like two cores. taks
  3. heck, some porn flicks actually have PLOT!!! mine come in and start whining every morning now that i'm the first one up. they know i'm the one that puts food in their bowl. the boy got himself a bath a day or two ago and he's been pretty bent out of shape over the matter. fat slob that one is. taks
  4. for your situation, that made reasonable sense other than the driver/compatibility issues. i got XP pro probably 2 years ago, well before vista was on anyone's radar. i never fell for the ME trap, btw. taks
  5. vista is too new... XP pro works fine and i don't have to worry about compatibility issues or lack of drivers. i don't have a DX10 vid card, either. that and i don't want to spend the money it will take to upgrade to vista when i have a perfectly legit copy of XP pro sitting at home (wife and son use XP home). taks
  6. totally custom, though not high-end. i typically rotate parts out giving the ejectees over to my wife's rig. i now have a son with a computer, so he'll join the musical pooter parts parade soon (his box is probably the strongest, minus my video card and HDD). the next "rotation" is going to happen soon. my wife complained that her CDs are no longer being recognized, and she's already had PATA controller issues... she'll get my stuff, and i'll get new. i'm still torn on AMD X2 or core2 duo. particularly with AMD about to release their core2 competition, which means high-end X2s will drop drastically in price, again. oh, and NO VISTA... taks
  7. por2 was completely playable unpatched... as long as you didn't install on a drive other than C. that it completely sucked is another matter. taks
  8. yeah, i've heard the biggest problems are in the toolset. those definitely fall into the category of bugs that should be fixed. however, even with such bugs that are apparent in every system, regardless of mitigating factors, diagnosis can be difficult. taks
  9. ALL software has bugs, and more particularly, software that is expected to run on multiple, unknown, platforms will be the worst of all. add that to the fact that winders is buggier 'n hell and you've got a great recipe for a trip down the amazon. really, what it comes down to is a question of percentage of players that experience various bugs, as well as those that are bothered by them. many of the games i've played in my life i've never experienced some of the most complained about bugs. other games, such as toee or por2, were a non-stop OFF-fest attempting to keep 'em at bay. just depends i spose. taks
  10. snipes are actual creatures, however. taks
  11. attribution is a very difficult problem. the goal is to find a correlation between a 2D object in an image, and a known 2D object. i work more in the detection realm, i.e. is the signal even there. both have their curiosities. taks
  12. what he was getting at is that after all the latencies are calculated in, the effect of slower memory is almost negligible in the long run. memory accesses are very bursty, and most applications are dominated by RAS/CAS latency (i.e. the hit you take every time you start an access) because it is rare for _all_ of the data you need to be located within a contiguous space in memory. the speedup from transferring data at a higher rate works out to almost nothing. as far as i know, any current processor can easily stream data at the rate it will come out of even the fastest memory, but that situation never arises in practice. taks
  13. this _used_ to be true, though not necessarily anymore. anandtech did a test with all the new processors and found that even budget RAM performed nearly as well as the high-end stuff in stock systems (i.e. no overclocking). the memory bus is so fast now, more is better than faster _unless_ you intend to overclock. in general, the bottleneck is going to be reads/writes to disk, so the more RAM you have, the less disk accesses you have. regard this as a very loose rule of thumb since the landscape of computer architecture has changed dramatically in the last few years. taks
  14. image processing is actually a branch of EE under the guise of statistical signal processing. in fact, i do image processing of a sort. of course, the "images" that result from radar returns are a little different than an optical image (same conceptually, however). fun stuff. taks
  15. any system is only as strong as its weakest component. taks
  16. desired to become: since i was 7, electrical engineer became: electrical engineer i always had this vision of myself in a white lab coat (not sure why), leading a design team of sorts. trade the lab coat for a dead shirt, that's pretty much what i do. taks
  17. that was on one of the encore channels the other night. i fell asleep before it started. any good? the first one or two were good to fair, with a decrease in quality as the sequels progressed. i watched the hunter again yesterday. not a bad flick. i did not like the way they made LT's character into a "badass wimp." he's sulky, i can understand, because he spent his life training others to kill, even though he ultimately came to the conclusion that it was immoral (i'm assuming that's what they were trying to convey). i.e. he's racked with guilt. OK, i buy that. however, something about the way they portrayed his character made it come across as wimpy in some respect. i can't put my finger on it. either way, i like tommy lee as well as benicio del toro (even though he was a little stiff in this one). i also watched spiderman 1 and 2 recently. peter parker is a bumbling idiot, and i got the same feel for his character as i did LT above. he's a badass that swings from building to building. his character shouldn't be tripping over himself in the street. at least with superman the ineptitude was an act (though overblown in my opinion). this may not be a fault of the movie in this case, however, as it may be written into the comic books that way as well (i don't follow them). either way, i didn't buy it. it became an annoyance (except when he "lost his powers," and it fit at that point). taks
  18. fortunately, for me at least, texture settings don't have much of an impact on frame rate with NWN2. "minimum settings," btw, typically also carry with them the notion of "slide show" behavior. taks
  19. interesting... level 30 is only the base class limit... higher levels possible with multiclassing. hmmmm... taks
  20. doubtful, but easy to check either way. if you have an ohmmeter, check the resistance of the power lines to ground with the supply unconnected. it will be low, maybe on the order of a few hundred milliohms even, but not zero. taks
  21. i should point out that "low conductivity" typically means the connection to the wall outlet is on the order of 1 Mohm (i.e. 1,000,000 ohms). this allows a slow transfer of static to ground. static damage results from a fast transfer, i.e. higher current due to low resistance. taks
  22. yes, i agree. one wouldn't want to sound imprecise. taks
  23. i bought only one copy... standard edition. taks
  24. so, what... you're nearly certain there's a good possibility that its rumored sales may be high? taks
  25. i wouldn't recommend working on a glass table, btw. glass tends to hold static, which can cause problems. typical electrical work benches actually have a static mat that has low conductivity, and is connected to a ground point (such as a wall outlet). the wrist strap is good, however. given your probably higher than normal humidity (at least, higher than CO), static probably isn't a problem anyway. particularly for assembled boards. static is a poorly understood phenomenon even with many of the seasoned engineers that i've worked with. they often wrap their boards in the bags, rather than putting the boards in the bags, which is a huge no-no. static bags are designed to hold static on the outside of the bag, so wrapping a board in the bag is worse than not using a bag at all... silly electrical engineers... taks
×
×
  • Create New...