It makes good sense. Some operations take longer than others. So adjust the clock period to suit
each instruction. Only works if you execute one instruction at a time. Otherwise I can't see why you are puzzled. If, as at present, you run at a fixed clock speed, some instructions are ready before others, which is inefficient. Of course in the real world, it makes sense to keep the clock speed fixed, which is my worry about asynchronous designs; sounds good on paper, doesn't translate into zillion gate designs.
Synchronous design (using a clock) makes it a lot easier to structure your design. It makes it into a big state machine in effect.
There's nothing to stop you changing the clock period on a cycle by cycle basis - slow instruction, use a longer clock period. There used to be an AMD chip to do just that for bitslice machines. Even with a fixed clock period, only the critical path needs a full clock period, some bits of logic have made their minds up well before a clock tick, so the bottlenecks you mention apply here as well.
It all falls apart on the assumption that the clock ticks at various points in your circuit are all happening at the same time; speed of light is finite, so you can understand why that's not true in the real world. (Clock skew)
It also falls apart because you can't enforce clocking on the entire universe - a signal coming into your system can (in theory) hang it up if it comes at just the right time, but there's maths to show that the likelihood is so small that it doesn't matter in practice. (Metastability)
That said, last time I looked into asynchronous design, not very deeply I admit, the cure looked worse than the disease. Which isn't to say it's a bad idea, maybe its time has come?
implementing unprecedented new features such as these vertex shaders, rather than improving more general stuff such as fillrate or transformation and lighting
Duh! Vertex shaders -are- a generalisation (whatever) of the fixed function T&L pipeline!
Follow this bizarre URL which you can get to from the nVIDIA homepage thus: NVIDIA.Com> Developer Relations> Whitepapers> Implementation of "Missing" Vertex Shader Instructions. Work through the examples, don't just skim the text, follow the code execution. Then you'll grok vertex shaders.
You have a point about GeForce3 and DX8 fixing the hardware specification and therefore making it harder for similar innovations or improvements from other manufacturers to be adopted, but Carmack already said something similar about pixel shaders in his recent GeForce3 articles.
Support chipmakers that are seeking to make everything run better, like ATI and PowerVR.
I'm a big fan of Imagination Technologies' PowerVR and tile based rendering in general just in case you think I'm an nVIDIA groupie.
All I see here is
another kneejerk/. conspiracy theory. Sigh.
I'm prepared to risk being moderated as offtopic, or being flamed for not trawling every last corner of the WWW but I've come across a case of the technology not quite surviving the cost reduction process...
I acquired a GeForce2MX PCI for an old homebox to see if the hardware T&L would make a difference with a tired old Winchip3D class processor - the box is so old is doesn't have AGP.
Thing is, the card doesn't even display the usual BIOS message, the (linux-only) box boots up OK but with a black screen. I've fiddled with zillions of BIOS settings, no luck at all. What has happened that breaks things so fundamentally? At least I learned about repairing the file system I accidentally trashed when I unthinkingly turned the box off to replace the old TNT board...
Anyone who'd shell out six hundred for one of these is insane.
So programmers are insane? $600 gets you time to develop your game ready for when such cards are more affordable. Or time to hone your skills to get a better job. Or just for the fun of messing with the card.
Maybe they've really only implemented the three least significant bits of each colour on the chip to save on die area? Explains why Doom3 is so dark doesn't it?
Mod him up for hitting the nail on the head. I just wish there was some way of proving his point, or at least of reducing the subjectivity, for example taking an arcade game and seeing how much effort is needed to port it to DC and PS2, and then asking users which looked and played better.
If you can give me a good URL or book to read, it will be very nice.
The work I did is history. I only hope that a company like nVIDIA will find a way to implement something like it now that millions of gates can be put on a consumer chip. Your reply suggests you are looking to the future - what I would like to see someday is real-time RenderMan, so I think you'd be interested in the Advanced RenderMan book by Apodaca and Gritz.
Civilised reply, thanks. Bad AA obscures detail, displays are bad. Some flight simulators do their very best to be accurate and faithful; people sneer at 1000 polygon systems from 1980, but when it comes to training pilots, civilian as well as military, they emphasise what matters.
I'm very reassured to know that commercial pilots must rain on simulators at regular intervals. If bad AA enforced bad flying habits, I'd be very worried, or dead. Ditto if bad AA caused military pilots to get trigger-happy or careless. QED.
You're right, memory bandwidth is the problem. Do a web search on tile based rendering (eg the PowerVR used in Dreamcast or the patent I mention elsewhere in this discussion). There's no need to waste memory bandwidth on anti-aliasing.
Wireframe doesn't describe the run-time intersections between polygons. Imagine a house on the ground. The ground isn't flat, you don't want to waste a zillion polygons making the bottom of the house match the terrain contours, even if you tried floating point errors would catch you out. Also, in the real world, bricks instersect the ground.
If you want to win a computer game played for entertainment, disable the anti-aliasing. When a target appears in the distance, it will pop and sparkle and generally draw attention to itself so you can aim at it and kill it.
If you are in the military and are using the computer as a way of practicing your skills without getting killed, killing others or wasting ammo, then enable anti-aliasing. The first hundred times you'll die (and insert a quarter to try again) but thereafter you'll know why it's so hard to spot a blip on the horizon approaching at mach two.
The anti-aliasing method discussed here detects edges in the model
Oh dear. If two polygons intersect each other at run-time, the line of intersection will look jaggy if all one does is anti-alias polygon edges in the model.
You're so right, brute force is the way to go, but there's dumb brute force and smart brute force...
Of course you have to use supersampling, duh. I'm dead serious, no need to be flippant with the comment about polarization and Wavelength effects. I've got the boards on my wall at work and a couple of dead chips as souvenirs. Read the patent.
Take a look at this patent. I designed an 11 million gate system (in 1990:-) to implement these guys' ideas. I can't discuss anything not in the public domain, but if anyone has any comments I'm able to answer, I'd be only too pleased, because this subject is one of my long term geek interests.
What matters is to solve three problems simultaneously, not only anti-aliasing but depth buffering and translucency.
I'm Ella the Cat. I wasn't bought, neither was my sister from the next litter. If no-one took us in, maybe we'd have been turned out on the streets of urban SE London, far from where I live now. I was the runt of a litter and AFAIK the last one alive. I've been sterilised after going on heat once, which, admittedly, I had no choice about. I could have gone off someplace else any time I wanted. I live on biscuits on vetinary advice.
I'm not owned and I'm not as dumb as you make me or my owner out to be. Being anthropomorphised is the least of my worries.
Go read post 49. Yours is post 50 and 3 minutes later so you could have missed it.
Let's see how well you argue complex moral issues. Moderators may flag this as offtopic, but if Slashdot is to keep its focus, news for nerds, stuff that matters, we have to stop debates going offtopic by challenging people who take them offtopic (mea culpa?). You think your view matters, I wouldn't doubt your commitment, so here goes.
Your post (#8) says SuSE should donate to enlightened causes, including animal rights groups. Ella is a pet cat. There are anecdotal stories of how experiments on new born kittens determined that the visual system is conditioned at a very early age, as a consequence operations to correct a squint in humans now take place almost immediately after birth, whereas before it was felt appropriate to wait a while.
As Hubel and Wiesel first observed in the 1960s 56, closing one eye in a kitten during its early
postnatal development profoundly disrupted the pattern of ODC; the eye that had visual input
dominated the cortex, whereas the eye that had been closed lost its connections.
I couldn't do this to a kitten, nor could I condemn a human to suffer a squint. My point is that issues like this are complex and subtle.
Slashdot suffers when people take simplistic moral positions that are tangential to the topic.
Just because an issue is worthy, doesn't mean you should introsuce it into every discussion.
I'm Ella the Cat. The human I get to do my typing who views me as a pet seems happy to pay SuSE for the convenience of getting a working system. As a cat, I'm pretty much laid back, but my human gets really tetchy when reading between the lines of posts like yours.
Don't worry about it. Someone gave me a karma point and took one off you, which made me feel guilty. Good to know you're not upset. I've only got 13 and don't care about my karma (New Year's resolution:-) as long as I can post at the default level. I just get tetchy when it looks like people are being rewarded with karma for -not- reading the main article.
Did anyone get the Ohno! joke btw? If you're moderating and think it was funny, give someone else who looks like they need it the points.
"Currently, transistors in laptop displays absorb a quarter of the brightness of the backlight. Transparent transistors could solve this."
Above quote copy-pasted from the New Scientist article referenced by/. - I claim score 3 informative too. No offence intended to onion2k, but this is getting silly.
BTW, If transistors soak up 25% of the light, just stick a few on your window, then the sun won't be so bright
Here's the website you're probably thinking of. I like the bit about single precision floats not being sufficient (yep, this is one of my bookmarks, not a 10 second rummage on Google for karma).
I see you're from the UK from your URL, as am I. I never say "legos" with an s, and I'd feel safe to say most people in the UK wouldn't either, but I've noticed a lot of people on/. do, so I'd
guess you're objecting to US usage. Vive la difference!
Slashdot Mirror
Will it be road tested?
What I love about that show is how robots built from elastic bands, cardboard and milk bottle tops manage to beat the big guys.
It makes good sense. Some operations take longer than others. So adjust the clock period to suit each instruction. Only works if you execute one instruction at a time. Otherwise I can't see why you are puzzled. If, as at present, you run at a fixed clock speed, some instructions are ready before others, which is inefficient. Of course in the real world, it makes sense to keep the clock speed fixed, which is my worry about asynchronous designs; sounds good on paper, doesn't translate into zillion gate designs.
Synchronous design (using a clock) makes it a lot easier to structure your design. It makes it into a big state machine in effect.
There's nothing to stop you changing the clock period on a cycle by cycle basis - slow instruction, use a longer clock period. There used to be an AMD chip to do just that for bitslice machines. Even with a fixed clock period, only the critical path needs a full clock period, some bits of logic have made their minds up well before a clock tick, so the bottlenecks you mention apply here as well.
It all falls apart on the assumption that the clock ticks at various points in your circuit are all happening at the same time; speed of light is finite, so you can understand why that's not true in the real world. (Clock skew)
It also falls apart because you can't enforce clocking on the entire universe - a signal coming into your system can (in theory) hang it up if it comes at just the right time, but there's maths to show that the likelihood is so small that it doesn't matter in practice. (Metastability)
That said, last time I looked into asynchronous design, not very deeply I admit, the cure looked worse than the disease. Which isn't to say it's a bad idea, maybe its time has come?
implementing unprecedented new features such as these vertex shaders, rather than improving more general stuff such as fillrate or transformation and lighting
Duh! Vertex shaders -are- a generalisation (whatever) of the fixed function T&L pipeline!
Follow this bizarre URL which you can get to from the nVIDIA homepage thus: NVIDIA.Com> Developer Relations> Whitepapers> Implementation of "Missing" Vertex Shader Instructions. Work through the examples, don't just skim the text, follow the code execution. Then you'll grok vertex shaders.
You have a point about GeForce3 and DX8 fixing the hardware specification and therefore making it harder for similar innovations or improvements from other manufacturers to be adopted, but Carmack already said something similar about pixel shaders in his recent GeForce3 articles.
Support chipmakers that are seeking to make everything run better, like ATI and PowerVR.
I'm a big fan of Imagination Technologies' PowerVR and tile based rendering in general just in case you think I'm an nVIDIA groupie.
All I see here is another kneejerk /. conspiracy theory. Sigh.
I'm prepared to risk being moderated as offtopic, or being flamed for not trawling every last corner of the WWW but I've come across a case of the technology not quite surviving the cost reduction process ...
I acquired a GeForce2MX PCI for an old homebox to see if the hardware T&L would make a difference with a tired old Winchip3D class processor - the box is so old is doesn't have AGP.
Thing is, the card doesn't even display the usual BIOS message, the (linux-only) box boots up OK but with a black screen. I've fiddled with zillions of BIOS settings, no luck at all. What has happened that breaks things so fundamentally? At least I learned about repairing the file system I accidentally trashed when I unthinkingly turned the box off to replace the old TNT board ...
Anyone who'd shell out six hundred for one of these is insane.
So programmers are insane? $600 gets you time to develop your game ready for when such cards are more affordable. Or time to hone your skills to get a better job. Or just for the fun of messing with the card.
I came across the Filesystem Hierarchy Standard rummaging around /usr/share/doc on my SuSE box.
Maybe they've really only implemented the three least significant bits of each colour on the chip to save on die area? Explains why Doom3 is so dark doesn't it?
Mod him up for hitting the nail on the head. I just wish there was some way of proving his point, or at least of reducing the subjectivity, for example taking an arcade game and seeing how much effort is needed to port it to DC and PS2, and then asking users which looked and played better.
fluffy_bunny = reinterpret_cast<CuddlyThing>(grumpy_wizard) ;
If you can give me a good URL or book to read, it will be very nice.
The work I did is history. I only hope that a company like nVIDIA will find a way to implement something like it now that millions of gates can be put on a consumer chip. Your reply suggests you are looking to the future - what I would like to see someday is real-time RenderMan, so I think you'd be interested in the Advanced RenderMan book by Apodaca and Gritz.
Civilised reply, thanks. Bad AA obscures detail, displays are bad. Some flight simulators do their very best to be accurate and faithful; people sneer at 1000 polygon systems from 1980, but when it comes to training pilots, civilian as well as military, they emphasise what matters.
I'm very reassured to know that commercial pilots must rain on simulators at regular intervals. If bad AA enforced bad flying habits, I'd be very worried, or dead. Ditto if bad AA caused military pilots to get trigger-happy or careless. QED.
You're right, memory bandwidth is the problem. Do a web search on tile based rendering (eg the PowerVR used in Dreamcast or the patent I mention elsewhere in this discussion). There's no need to waste memory bandwidth on anti-aliasing.
Wireframe doesn't describe the run-time intersections between polygons. Imagine a house on the ground. The ground isn't flat, you don't want to waste a zillion polygons making the bottom of the house match the terrain contours, even if you tried floating point errors would catch you out. Also, in the real world, bricks instersect the ground.
If you want to win a computer game played for entertainment, disable the anti-aliasing. When a target appears in the distance, it will pop and sparkle and generally draw attention to itself so you can aim at it and kill it.
If you are in the military and are using the computer as a way of practicing your skills without getting killed, killing others or wasting ammo, then enable anti-aliasing. The first hundred times you'll die (and insert a quarter to try again) but thereafter you'll know why it's so hard to spot a blip on the horizon approaching at mach two.
The anti-aliasing method discussed here detects edges in the model
Oh dear. If two polygons intersect each other at run-time, the line of intersection will look jaggy if all one does is anti-alias polygon edges in the model.
You're so right, brute force is the way to go, but there's dumb brute force and smart brute force ...
Of course you have to use supersampling, duh. I'm dead serious, no need to be flippant with the comment about polarization and Wavelength effects. I've got the boards on my wall at work and a couple of dead chips as souvenirs. Read the patent.
Not everything on /. is a troll or done for karma.
Take a look at this patent. I designed an 11 million gate system (in 1990:-) to implement these guys' ideas. I can't discuss anything not in the public domain, but if anyone has any comments I'm able to answer, I'd be only too pleased, because this subject is one of my long term geek interests.
What matters is to solve three problems simultaneously, not only anti-aliasing but depth buffering and translucency.
I'm Ella the Cat. I wasn't bought, neither was my sister from the next litter. If no-one took us in, maybe we'd have been turned out on the streets of urban SE London, far from where I live now. I was the runt of a litter and AFAIK the last one alive. I've been sterilised after going on heat once, which, admittedly, I had no choice about. I could have gone off someplace else any time I wanted. I live on biscuits on vetinary advice.
I'm not owned and I'm not as dumb as you make me or my owner out to be. Being anthropomorphised is the least of my worries.
Go read post 49. Yours is post 50 and 3 minutes later so you could have missed it.
Let's see how well you argue complex moral issues. Moderators may flag this as offtopic, but if Slashdot is to keep its focus, news for nerds, stuff that matters, we have to stop debates going offtopic by challenging people who take them offtopic (mea culpa?). You think your view matters, I wouldn't doubt your commitment, so here goes.
Your post (#8) says SuSE should donate to enlightened causes, including animal rights groups. Ella is a pet cat. There are anecdotal stories of how experiments on new born kittens determined that the visual system is conditioned at a very early age, as a consequence operations to correct a squint in humans now take place almost immediately after birth, whereas before it was felt appropriate to wait a while.
I can't verify the anecdote, but I did some digging around, and I came across this
As Hubel and Wiesel first observed in the 1960s 56, closing one eye in a kitten during its early postnatal development profoundly disrupted the pattern of ODC; the eye that had visual input dominated the cortex, whereas the eye that had been closed lost its connections.
I couldn't do this to a kitten, nor could I condemn a human to suffer a squint. My point is that issues like this are complex and subtle.
Slashdot suffers when people take simplistic moral positions that are tangential to the topic. Just because an issue is worthy, doesn't mean you should introsuce it into every discussion.
I'm Ella the Cat. The human I get to do my typing who views me as a pet seems happy to pay SuSE for the convenience of getting a working system. As a cat, I'm pretty much laid back, but my human gets really tetchy when reading between the lines of posts like yours.
Don't worry about it. Someone gave me a karma point and took one off you, which made me feel guilty. Good to know you're not upset. I've only got 13 and don't care about my karma (New Year's resolution :-) as long as I can post at the default level. I just get tetchy when it looks like people are being rewarded with karma for -not- reading the main article.
Did anyone get the Ohno! joke btw? If you're moderating and think it was funny, give someone else who looks like they need it the points.
"Currently, transistors in laptop displays absorb a quarter of the brightness of the backlight. Transparent transistors could solve this."
Above quote copy-pasted from the New Scientist article referenced by /. - I claim score 3 informative too. No offence intended to onion2k, but this is getting silly.
BTW, If transistors soak up 25% of the light, just stick a few on your window, then the sun won't be so bright
Here's the website you're probably thinking of. I like the bit about single precision floats not being sufficient (yep, this is one of my bookmarks, not a 10 second rummage on Google for karma).
I see you're from the UK from your URL, as am I. I never say "legos" with an s, and I'd feel safe to say most people in the UK wouldn't either, but I've noticed a lot of people on /. do, so I'd
guess you're objecting to US usage. Vive la difference!