... since modern GPUs have much less application-specific logic... this may be the trend on a relative scale, in that the percentage of logic gates that are used for programmable hardware is increasing. however, an an absolute scale, the fixed-function logic (things like triangle setup, MSAA resolve, etc) is certainly using more logic gates with every new generation of parts. i'm not really disagreeing with you, just noting that it all depends on how you look at it.
Well, yeah, and if I say the context is the motherboard, then I can define a four-socket board holding single-core CPUs to be a quad core chip. It would be equally ludicrous in the context of chip design.
ignoring the cores-per-die discussion, i find it interesting that you view asic (eg chip) design to stop at the die level. package design and development is a *huge* part of designing and manufacturing an asic. the two parts are designed together, and i find it difficult to view them as completely separate entities.
generally speaking, you're quite right (so i'm not disagreeing here, just offering an anecdote...).
while debugging an issue on one of our graphics boards, i found that a number of samsung DRAMs (some variety of gddr1 parts) would hold their contents for a good 15+ seconds without any accesses or refresh cycles. how do i know? i disabled the refresh cycles, disabled the display-refresh requests (debugging via serial console on an adjacent dos box), and had a scope and logic analyzer hooked up. the only memory accesses were my directed read/write cycles. the part wasn't in self-refresh mode either... and i could consistently read back the arbitrary data that i wrote to the framebuffer 15 seconds prior, with no way for the part to refresh that particular row (or any other), save my subsequent read. it was pretty neat, albeit fairly pointless.
and no, it didn't really work quite so well with a power-cycle in between accesses. some bits did seem to retain their state for a while (5+ seconds), but reading both bytes (x16 parts) wasn't very reliable.
the thing is, rarely are clock trees built as completely balanced structures... usually some sort of clock-tree-synthesis tool is used, and while the tree may be balanced for a particular PVT, care has to be taken to balance transistor delay with metal delay, across the entire structure. hold fixing is a major part of timing closure on an asic.
but in general, i agree with your point, the faster the chip runs (not clock frequency but signal delay time), eventually you'll get hold violations.
indeed, i always use 'dice' at the office (EE working at fab-less semiconductor outfit mentioned here often enough...), and tend to cringe a bit when i hear or read 'dies'. generally speaking though, any of 'dice', 'dies', or 'die' seems to be a generally acceptable plural form of 'die', to my occaisional chagrin.
for me, it makes sense to dice a wafer and end up with, well, dice. dicing and ending up with dies doesn't quite sound right. and in manufacturing (excluding wafer processing for the sake of argument), the die is the tool, not the product. i guess you could take an IC die and try to use it to stamp or cast something if you really wanted to though.
"My stance is, I don't know if this is illegal, immoral or unconstitutional, thats the Court's role."
i would suggest that most anyone should be able to determine whether something is immoral or not. whether others, let alone a majority, agree with you is tangential. as to whether something is illegal or unconstitutional, again, i think most people/citizens should be able to make such a determination for themselves. the legislature or court may not agree, but then it becomes a question of how well you're represented by your government, and what (if anything) you're going to do about it. of course, we're all free to absolve ourselves of the right (responsibility?) to make these determinations for ourselves, but i don't view that as being a particularly good idea.
generally speaking, i tend to agree with you on this particular issue. one analogy that i would make is to relate these consoles to automobiles. once bought, i'm free to do whatever i like with my car. and if i reprogram the ECU or change the suspension, or even replace the whole engine, the car still works. the manufacturers sometimes make it so fancy expensive tools are required to do some very basic maintenance, but i can still do whatever i want to the car i bought, and have it still operate fine. and on the other hand, if i'm leasing it, then i can't really tinker however i'd like, since the car isn't really mine.
on the console front, the business model is selling me the console, but in many cases, effectively operating as if the box is under a lease agreement. if i've bought the machine, and the system software installed on it, there's no reason i shouldn't be able to do whatever i like with the system. the fact that i can't, whatever the rationale, is bothersome.
"People will be clamoring over the fact 'the Man is invading my privacy!' and then beg for more security and question how 'we let so many terrorists get by'."
by and large, i don't think the people uttering those two phrases are going to be the same...
that's an interesting take on the argument. in all seriousness, do you really read capitalized words as imparting some sort of intonation onto the sentence or paragraph?
one thing that i can maybe see wanting is capital letters at the start of a sentence, to help delineate them. but i always put two spaces between sentences for that very purpose (even when typing official documentation wherein proper capitalization is required). since the html convention is to compress whitespace to the minimum required for word-separation, sentences which don't begin with a capital letter may tend to look like a run-on. to my mind though, capitalization doesn't really help much there.
in anyone speaking english, i've never heard cache pronounced with a long-a sound (eg cage). mispronunciations annoy me to no end (library as lie-berry is an egregious example), but at the same time, regional accents fascinate me.
anyway, i'm wondering where you've heard the long-a pronunciation. was it really listed that way in the dictionary?
also, the 'u' in flush is usually (in american english, at any rate) pronounced the same as the 'a' in abut (or the sound "uh"). i think that sounds is called a schwa. and the part of me that speaks french would pronounce the 'a' in cacher as more of an "ahh" sound. definately different sounds to my ear.
anyway, those are my two random cents on this subject...
while i generally agree, i have to take some exception to the capitalization bit. i find the shift keys (both of them) are rather inconvenient to hit while typing (yes, i'm rather certain my technique would make Mavis Beacon cringe, but i can do it quite fast and accurately), so i almost never capitalize anything words other than names of people (other proper nouns have to live without). it's not about being contrary, simply convenience. and in this case, convenience means avoiding awkward finger stretches for the sake of speed and digital comfort.
further, this habit i've developed is reinforced by working almost exclusively in a computer environment wherein capital letters are infrequently used (eg unix). but, other than my aversion to the shift key, i try to spell correctly and use proper grammar. when that doesn't happen, it's almost always a case of my mind or fingers getting ahead of the other, and i cringe when i read such mistakes in my own writing.
still, i suppose the capitalization issue bothers some as much as missing commas and malapropisms tend to bother me. to each his own, n'est pas?
"I'm not sure I'd want to be able to condem her to death to save the aircraft. I'm not sure I'd want someone with that level of detachment flying my plane."
so you'd prefer to have someone willing and able to hold a razor to her throat to control the plane instead...?
Slashdot Mirror
indeed, the government never does anything illegal...
i always think "equivalent series resistance" initially, and often find myself confused for a moment when reading those three letters on slashdot...
forgive the non sequitur, but all else aside, there's at least one good thing about the south... Mott's pit-cooked BBQ... mmm...
... since modern GPUs have much less application-specific logic...this may be the trend on a relative scale, in that the percentage of logic gates that are used for programmable hardware is increasing. however, an an absolute scale, the fixed-function logic (things like triangle setup, MSAA resolve, etc) is certainly using more logic gates with every new generation of parts. i'm not really disagreeing with you, just noting that it all depends on how you look at it.
on-die MOS capacitors are actually quite common.
naja, it was a tongue-in-cheek jab... i followed your point, and generally agree with it. i just found your initial phrase semantically amusing...
err... the song "In Other Words" was written/composed a good 40+ years before evangelion was produced...
"Such wonders aren't exactly wonderous..."
i'm not quite sure how to parse this...
Fanboy, zealot, fanatic...
red sox fan...
generally speaking, you're quite right (so i'm not disagreeing here, just offering an anecdote...).
while debugging an issue on one of our graphics boards, i found that a number of samsung DRAMs (some variety of gddr1 parts) would hold their contents for a good 15+ seconds without any accesses or refresh cycles. how do i know? i disabled the refresh cycles, disabled the display-refresh requests (debugging via serial console on an adjacent dos box), and had a scope and logic analyzer hooked up. the only memory accesses were my directed read/write cycles. the part wasn't in self-refresh mode either... and i could consistently read back the arbitrary data that i wrote to the framebuffer 15 seconds prior, with no way for the part to refresh that particular row (or any other), save my subsequent read. it was pretty neat, albeit fairly pointless.
and no, it didn't really work quite so well with a power-cycle in between accesses. some bits did seem to retain their state for a while (5+ seconds), but reading both bytes (x16 parts) wasn't very reliable.
"I've seen a sweet little girl, 8 years old, beat up her grandma..."
this is a rather intesting use of the adjective "sweet"...
errr... by "signal delay time" i meant combinational (eg transistor) delay. that's what i get for trying to think on my day off...
btw, when did this successive-submit delay show up? it's kind of annoying...
the thing is, rarely are clock trees built as completely balanced structures... usually some sort of clock-tree-synthesis tool is used, and while the tree may be balanced for a particular PVT, care has to be taken to balance transistor delay with metal delay, across the entire structure. hold fixing is a major part of timing closure on an asic.
but in general, i agree with your point, the faster the chip runs (not clock frequency but signal delay time), eventually you'll get hold violations.
indeed, i always use 'dice' at the office (EE working at fab-less semiconductor outfit mentioned here often enough...), and tend to cringe a bit when i hear or read 'dies'. generally speaking though, any of 'dice', 'dies', or 'die' seems to be a generally acceptable plural form of 'die', to my occaisional chagrin.
for me, it makes sense to dice a wafer and end up with, well, dice. dicing and ending up with dies doesn't quite sound right. and in manufacturing (excluding wafer processing for the sake of argument), the die is the tool, not the product. i guess you could take an IC die and try to use it to stamp or cast something if you really wanted to though.
supposing that's true, one might also suppose that said soldiers and pilots volunteered for that training. interesting distinction, i think.
holy crap, yes! the underwater wall bit was pretty frustrating, but the spiked columns were 100 times worse.
"My stance is, I don't know if this is illegal, immoral or unconstitutional, thats the Court's role."
i would suggest that most anyone should be able to determine whether something is immoral or not. whether others, let alone a majority, agree with you is tangential. as to whether something is illegal or unconstitutional, again, i think most people/citizens should be able to make such a determination for themselves. the legislature or court may not agree, but then it becomes a question of how well you're represented by your government, and what (if anything) you're going to do about it. of course, we're all free to absolve ourselves of the right (responsibility?) to make these determinations for ourselves, but i don't view that as being a particularly good idea.
generally speaking, i tend to agree with you on this particular issue. one analogy that i would make is to relate these consoles to automobiles. once bought, i'm free to do whatever i like with my car. and if i reprogram the ECU or change the suspension, or even replace the whole engine, the car still works. the manufacturers sometimes make it so fancy expensive tools are required to do some very basic maintenance, but i can still do whatever i want to the car i bought, and have it still operate fine. and on the other hand, if i'm leasing it, then i can't really tinker however i'd like, since the car isn't really mine.
on the console front, the business model is selling me the console, but in many cases, effectively operating as if the box is under a lease agreement. if i've bought the machine, and the system software installed on it, there's no reason i shouldn't be able to do whatever i like with the system. the fact that i can't, whatever the rationale, is bothersome.
"People will be clamoring over the fact 'the Man is invading my privacy!' and then beg for more security and question how 'we let so many terrorists get by'."
by and large, i don't think the people uttering those two phrases are going to be the same...
"...or make people pay into their own private funds."
is it really so terrible to expect people to save for their own futures?
that's an interesting take on the argument. in all seriousness, do you really read capitalized words as imparting some sort of intonation onto the sentence or paragraph?
one thing that i can maybe see wanting is capital letters at the start of a sentence, to help delineate them. but i always put two spaces between sentences for that very purpose (even when typing official documentation wherein proper capitalization is required). since the html convention is to compress whitespace to the minimum required for word-separation, sentences which don't begin with a capital letter may tend to look like a run-on. to my mind though, capitalization doesn't really help much there.
anyway, happy to bore you...
speaking of pronunciation...
in anyone speaking english, i've never heard cache pronounced with a long-a sound (eg cage). mispronunciations annoy me to no end (library as lie-berry is an egregious example), but at the same time, regional accents fascinate me.
anyway, i'm wondering where you've heard the long-a pronunciation. was it really listed that way in the dictionary?
also, the 'u' in flush is usually (in american english, at any rate) pronounced the same as the 'a' in abut (or the sound "uh"). i think that sounds is called a schwa. and the part of me that speaks french would pronounce the 'a' in cacher as more of an "ahh" sound. definately different sounds to my ear.
anyway, those are my two random cents on this subject...
while i generally agree, i have to take some exception to the capitalization bit. i find the shift keys (both of them) are rather inconvenient to hit while typing (yes, i'm rather certain my technique would make Mavis Beacon cringe, but i can do it quite fast and accurately), so i almost never capitalize anything words other than names of people (other proper nouns have to live without). it's not about being contrary, simply convenience. and in this case, convenience means avoiding awkward finger stretches for the sake of speed and digital comfort.
further, this habit i've developed is reinforced by working almost exclusively in a computer environment wherein capital letters are infrequently used (eg unix). but, other than my aversion to the shift key, i try to spell correctly and use proper grammar. when that doesn't happen, it's almost always a case of my mind or fingers getting ahead of the other, and i cringe when i read such mistakes in my own writing.
still, i suppose the capitalization issue bothers some as much as missing commas and malapropisms tend to bother me. to each his own, n'est pas?
"I'm not sure I'd want to be able to condem her to death to save the aircraft. I'm not sure I'd want someone with that level of detachment flying my plane."
so you'd prefer to have someone willing and able to hold a razor to her throat to control the plane instead...?