Binary translation is a black art, Transmeta's know-how plus their patents on hardware assisted x86 emulation might be the ticket AMD wants to finally put the whole ugly mess x86 has become behind them without giving up backwards compatibility.
So theres no real problem... I think microsoft employ's some of the best engineers and scientists in the world, and I still think they produce a lot of dreck (and a lot of good stuff too).
Why would respect for a couple of group members mean you could not critisize the group as a whole and their works?
For a large amount of people thats not true, CT did a nice blind test (I know I know, a blind test in high end audio testing... whats the world coming to, everyone knows subjective tests only count if you know what you are listening to ay?). There were only very few songs in which a majority of the people could distuingish 256Kb/s from the original... and this is just a transparancy test, not a quality one. In blind quality testing even lower bitrates would be enough.
VQ is used in a lot more area's than graphics, and in general it just means that you approximate multidimensional vectors by taking one set of vectors (the codebook) and approximate all the others by whichever one in the set closest to them.
If 3dfx is using the term to describe their S3TC (which was itself an offshoot of a small part of MPEG-2 BTW) offshoot they are bastardizing the term, it wouldnt be the first time.
Its meant to render a 256^3 voxel set in real time with full transparancy, not to quickly render a displacement mapped surface... which I agree is very usefull and has bugger all to do with voxels.
If you consider somoene at whatever website giving his opinion with 0 arguments to back them up as a political evaluation to be taken seriously its only a small step to accepting Taco as your oracle.
He just gave his opinion, dont make more of it than it is.
Trademark squatting, this has happened before. Its no accident, it happened for the word pentium too they tried to hussle money out of anyone who put it in add's.
featuring a reduced core size, lower power requirements and large, full-speed, on-die L2 cache"
Ill take a _literal_ quote, which you conveniently didnt bother to put in your post, over grasping deductions with AFAICS no real basis but your own intuition thanks.
Willamette is not specifically a 3D powerhouse IMO, it doesnt do any more flops per cycle, its a clock monster... like any modern CPU (except PowerPC/G4, but then... thats a bit of a dog). If Mustang cant keep up in clock its fucked, if they can their market share will keep groing.
Articles are nice, but with no benchmarks or even inside information to back them up they are just so much hot air.
"PS> It isn't possible to change the process without changing the layout, but steppings almost never result in performance increases, just better yeilds. Case in point. A.35 micron PII overclocked to 333 gives the exact same performance as a 333 MHz.25 micron PII." I agree, but then thats the only thing AMD has to do to be able to keep up... get high yields in high frequency bins.
What they arent saying is that the next step is legislation to enforce that any new system capable of digital sound reproduction made/imported has to check for these watermarks.
I dont see how they wont lobby for this, the same has happened for analog video... so dont think it cant happen. Better be carefull with your portable mp3 players, they wont be making em like that for long.
Actually no, I dont expect either Intel or AMD to notify us each and every time they change the layout of the chip a bit to remove timing bottlenecks... they just change the stepping of the chip update the errata if necessary and move on with shipping faster processors. Why would people care exactly why processors get faster as long as they do? The pipeline doesnt have to change, although it can happen like for instance the longer reorder buffer introduced but never publicized early in the Athlon's life, and neither the feature set. From a software point of view nothing changes but the clock.
And you cant just shrink a design from.35u.25u its never that easy...
"Don't you think they would put it in the FAQ is the actual chip was going to be changed" No I dont since they didnt publicize the reorder buffer thing either, but they put it in anyway... as I see it you just choose to read it as if they didnt for some bizarre reason.
Here another AMD quote: "Mustang is planned to be an enhanced version of Thunderbird, featuring a reduced core size, lower power requirements and large, full-speed, on-die L2 cache. Multiple versions of the Mustang core are planned to be targeted at the high-performance server/workstation, and high-performance desktop, and mobile markets. As for frequencies, we think we can offer processors that operate at competitive frequencies."
Specific enough for you? Or are they just outright lying here huh?
( from http://www.3dgn.com/hardware/articles/amd/index.sh tml )
Yes it is a nice article, it tries to make very clear what parts are guesswork:) In essence the main difference with the P4 is that it can potentially lower latency in respect to PIII for the SSE instructions, because it doesnt have to do two passes through the 64 bit pipes, throughput is unaffected... wether it actually does that, I have my doubts. And Im too lazy to check:) As you said yourself though, latency is not much of a problem for 3D:)
You still have a strange concept of proof... noone without so many preconceptions as you would read AMD's FAQ like that let alone call it proof of them. You cant just move the chip to another process (copper) without making some changes.
Unlike the Willamette with SSE K6-2 and K7 can execute 2 3DNow! instructions per cycle:) (although the K6-2 cant fully benefit from that due to other bottlenecks) 4*1=2*2
As for Mustang, Ill just quote the official Athlon FAQ from AMD...
"A15: "Mustang"
Enhanced version of AMD Athlon processor with reduced core size, lower power requirements, and up to 1MB of on-chip, performance-enhancing L2 cache memory. Manufactured on a 0.18 micron copper process technology. Multiple derivatives of the Mustang core are planned to address the requirements of the high-performance server/workstation, value/performance desktop and mobile markets."
The Willamette SDRAM chipset wont come out till next year according to that zdnet article you mention. Unlike anything in MaximumPC Intel has actually been quoted saying its aimed at the high end, you have a strange concept of proof.
Well you seem sincere to me, but then Im quite naive.
But you should try to be a little less adamant in your posts unless you are really truly totally sure of your case.
As for the Willamette thread... you perpetuated the myth of the two instruction per cycle SSE throughput (second pipeline doesnt do floating point). Intel quite clearly has yield problem in the high speed bins, they have admitted themselves the quantities are low. And wether or not AMD has anything to counter the Willamette remains to be seen, they have little problems pushing the clock on the Athlon/TB and the Mustang core has been tweaked that should help. As for the target market argument for the Mustang, the Willamette isnt targeted at the consumer space either. If AMD wants to they can migrate the Mustang core to its smaller cached brethren.
I like to be able to use windows drivers when I so wish, especially graphic card one's... X doesnt use half the potential of graphic cards.
Also DX is rushing ahead at mad cap speeds leaving Linux behind IMO, if you had an open source OS which could just use windows drivers to get acceleration to the DX API it would be a real alternative to windows for game porting. Unlike now where its becoming increasingly difficult for 3D game clients to be ported to Linux.
I noticed the incredible load of total BS in the AMD thread about the Willamette from you which was moderated up and now again here? Again with total bullshit. Do you have some agenda prooving something about the slashdot moderation system, or are you really sincere and just have flare for getting +1's out of clueless slashdot readers?
You can make hybrid wavelet/fractal compression schemes, but thats not the point. Wavelet compression in itself has fuck all to do with fractals.
Its seems pretty much essential for really low bandwith video.
3D wavelet compression has already been tried with IEZW and SPIHT without motion compensation, the paper for SPIHT merely compare's it to high bitrate MPEG-2... its slightly better but not dramatically so.
"However, at least some semblance of quality American/Western culture is being preserved by the numerous great and recognizeable literary artists we still have alive today"
Ill go along with that.
"Unfortunately such is not the case with modern Japanese culture"
Lets assume thats true for the moment you are going to present the evidence shortly, here it comes...
"For example, the only recent (in the late 20th century) great Japanese cultural celebrity who Americans have been mass exposed to is Akiro Kurasawa. Sadly, commercialized anime is being mass releaed dominating the exposure of true Japanese art and literature. Soon it seems as if true Japanese literature will become extinct as we are losing our connection to their culture. We should be preserving the true art of Japan, not killing it by buying up cheesy anime."
Oh wait I was mistaken, no proof present... all you have shown that the American perception of Japanese culture (according to your standards of course) is in decline. How US centric can you get? You arent being exposed to the creme the la creme of Japanese culture so it might as well not exist????
The Japanese are perfectly able of preserving those parts of their own culture they wish just like you are. Not too many Europeans read any of those great contemporary literary artists of your's either but we sure do mass consume your more popular entertainment. There is little difference between our relationship towards the US than your's with Japan apart from the fact that a little more of the artsy fartsy stuff can get accross due to the absense of the huge language barrier and because the large amounts of popular entertainment and common cultural background has given us a frame of reference, I doubt its because of their higher inherent quality.
How much story do you expect in an hour or so long film? Its just meant as entertainment, not art. Two of those stories are just meant as a small part of their bigger manga Universe anyway. And what the hell is wrong with titilation for its own sake? Are you a prude?
I am not American, so I wonder... That Albees and Updikes you mention who are saving you from cultural death, what percentage of Americans do you think which have voluntarily read them? I wouldnt be surprised if they arent "saving" a whole lot of Americans.
Not that people like me really want to be saved... there is a very good reason we dont seek out translated Basho, it would bore us to death. Its just like with Classical music, apart from the more populist stuff I just dont grok it... doesnt do a thing for me, if thats for a lack of education fine so be it. But I can still enjoy a good movie like the Matrix, or a good anime like Mahou Tsukai Tai or a good Sci-Fi novel like Snow-Crash (actually something like "Use of weapons" is more my thing, but I wanted to pick something even someoene has highly cultured as yourself was likely to recognise). Those suit my coarse taste just fine. In the end this is all about taste, you have yours... keep it.
Incompatible with what exactly? Not with BSD, thats for sure.
As for GPL everything is incompatible with it, if you believe the zealots anyway, the so called compatible licenses merely allow sublicensing... in themselves they are not compatible, the code has to change license.
Interbase's license will play with any license which isnt viral: "You may create a Larger Work by combining Covered Code with other code not governed by the terms of this License and distribute the Larger Work as a single product. In such a case, You must make sure the requirements of this License are fulfilled for the Covered Code."
I think the GPL has a questionable wording in the paragraph which makes it viral BTW... "You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program" linking to an unmodified library is not modification in my book.
A shared bus approach with 16 CPU's would not provide enough bandwith... and NUMA just lets programmers use bad code:)
IMO this machine is much too difficult to master for anyone but the tool developers, and since they are already dealing with so much nitty gritty dealing with the segmented memory directly (and more efficiently) wont add that much more work.
Unless you are talking about numbers for a specific benchmark (Linpack for instance) peak for a 1 GHz Athlon would be 4 GFLOP's using 3DNow!. Thats single precision, but then so is what the emotion engine uses.
Slashdot Mirror
Binary translation is a black art, Transmeta's know-how plus their patents on hardware assisted x86 emulation might be the ticket AMD wants to finally put the whole ugly mess x86 has become behind them without giving up backwards compatibility.
So theres no real problem... I think microsoft employ's some of the best engineers and scientists in the world, and I still think they produce a lot of dreck (and a lot of good stuff too).
Why would respect for a couple of group members mean you could not critisize the group as a whole and their works?
For a large amount of people thats not true, CT did a nice blind test (I know I know, a blind test in high end audio testing... whats the world coming to, everyone knows subjective tests only count if you know what you are listening to ay?). There were only very few songs in which a majority of the people could distuingish 256Kb/s from the original... and this is just a transparancy test, not a quality one. In blind quality testing even lower bitrates would be enough.
VQ is used in a lot more area's than graphics, and in general it just means that you approximate multidimensional vectors by taking one set of vectors (the codebook) and approximate all the others by whichever one in the set closest to them.
If 3dfx is using the term to describe their S3TC (which was itself an offshoot of a small part of MPEG-2 BTW) offshoot they are bastardizing the term, it wouldnt be the first time.
Its meant to render a 256^3 voxel set in real time with full transparancy, not to quickly render a displacement mapped surface... which I agree is very usefull and has bugger all to do with voxels.
This chip has near zero relevance to gaming.
Basically for everything but the landscape.
But it used bumpmapping and anti-aliasing so those looked quite nice for software too.
I think you are confusing CLI with a pure text console.
If you consider somoene at whatever website giving his opinion with 0 arguments to back them up as a political evaluation to be taken seriously its only a small step to accepting Taco as your oracle.
He just gave his opinion, dont make more of it than it is.
Trademark squatting, this has happened before. Its no accident, it happened for the word pentium too they tried to hussle money out of anyone who put it in add's.
featuring a reduced core size, lower power requirements and large, full-speed, on-die L2 cache"
.35 micron PII overclocked to 333 gives the exact same performance as a 333 MHz .25 micron PII."
Ill take a _literal_ quote, which you conveniently didnt bother to put in your post, over grasping deductions with AFAICS no real basis but your own intuition thanks.
Willamette is not specifically a 3D powerhouse IMO, it doesnt do any more flops per cycle, its a clock monster... like any modern CPU (except PowerPC/G4, but then... thats a bit of a dog). If Mustang cant keep up in clock its fucked, if they can their market share will keep groing.
Articles are nice, but with no benchmarks or even inside information to back them up they are just so much hot air.
"PS> It isn't possible to change the process without changing the layout, but steppings almost never result in performance increases, just better yeilds. Case in point. A
I agree, but then thats the only thing AMD has to do to be able to keep up... get high yields in high frequency bins.
What they arent saying is that the next step is legislation to enforce that any new system capable of digital sound reproduction made/imported has to check for these watermarks.
I dont see how they wont lobby for this, the same has happened for analog video... so dont think it cant happen. Better be carefull with your portable mp3 players, they wont be making em like that for long.
Actually no, I dont expect either Intel or AMD to notify us each and every time they change the layout of the chip a bit to remove timing bottlenecks... they just change the stepping of the chip update the errata if necessary and move on with shipping faster processors. Why would people care exactly why processors get faster as long as they do? The pipeline doesnt have to change, although it can happen like for instance the longer reorder buffer introduced but never publicized early in the Athlon's life, and neither the feature set. From a software point of view nothing changes but the clock.
.35u .25u its never that easy...
h tml )
And you cant just shrink a design from
"Don't you think they would put it in the FAQ is the actual chip was going to be changed" No I dont since they didnt publicize the reorder buffer thing either, but they put it in anyway... as I see it you just choose to read it as if they didnt for some bizarre reason.
Here another AMD quote:
"Mustang is planned to be an enhanced version of Thunderbird, featuring a reduced core size, lower power requirements and large, full-speed, on-die L2 cache. Multiple versions of the Mustang core are planned to be targeted at the high-performance server/workstation, and high-performance desktop, and mobile markets. As for frequencies, we think we can offer processors that operate at competitive frequencies."
Specific enough for you? Or are they just outright lying here huh?
( from http://www.3dgn.com/hardware/articles/amd/index.s
You force the term crackers on others too, how is that different?
Yes it is a nice article, it tries to make very clear what parts are guesswork :) In essence the main difference with the P4 is that it can potentially lower latency in respect to PIII for the SSE instructions, because it doesnt have to do two passes through the 64 bit pipes, throughput is unaffected... wether it actually does that, I have my doubts. And Im too lazy to check :) As you said yourself though, latency is not much of a problem for 3D :)
You still have a strange concept of proof... noone without so many preconceptions as you would read AMD's FAQ like that let alone call it proof of them. You cant just move the chip to another process (copper) without making some changes.
Unlike the Willamette with SSE K6-2 and K7 can execute 2 3DNow! instructions per cycle :) (although the K6-2 cant fully benefit from that due to other bottlenecks) 4*1=2*2
As for Mustang, Ill just quote the official Athlon FAQ from AMD...
"A15:
"Mustang"
Enhanced version of AMD Athlon processor with reduced core size, lower power requirements, and up to 1MB of on-chip, performance-enhancing L2 cache memory. Manufactured on a 0.18 micron copper process technology. Multiple derivatives of the Mustang core are planned to address the requirements of the high-performance server/workstation, value/performance desktop and mobile markets."
The Willamette SDRAM chipset wont come out till next year according to that zdnet article you mention. Unlike anything in MaximumPC Intel has actually been quoted saying its aimed at the high end, you have a strange concept of proof.
Well you seem sincere to me, but then Im quite naive.
But you should try to be a little less adamant in your posts unless you are really truly totally sure of your case.
As for the Willamette thread... you perpetuated the myth of the two instruction per cycle SSE throughput (second pipeline doesnt do floating point). Intel quite clearly has yield problem in the high speed bins, they have admitted themselves the quantities are low. And wether or not AMD has anything to counter the Willamette remains to be seen, they have little problems pushing the clock on the Athlon/TB and the Mustang core has been tweaked that should help. As for the target market argument for the Mustang, the Willamette isnt targeted at the consumer space either. If AMD wants to they can migrate the Mustang core to its smaller cached brethren.
I like to be able to use windows drivers when I so wish, especially graphic card one's... X doesnt use half the potential of graphic cards.
Also DX is rushing ahead at mad cap speeds leaving Linux behind IMO, if you had an open source OS which could just use windows drivers to get acceleration to the DX API it would be a real alternative to windows for game porting. Unlike now where its becoming increasingly difficult for 3D game clients to be ported to Linux.
I noticed the incredible load of total BS in the AMD thread about the Willamette from you which was moderated up and now again here? Again with total bullshit. Do you have some agenda prooving something about the slashdot moderation system, or are you really sincere and just have flare for getting +1's out of clueless slashdot readers?
You can make hybrid wavelet/fractal compression schemes, but thats not the point. Wavelet compression in itself has fuck all to do with fractals.
Its seems pretty much essential for really low bandwith video.
3D wavelet compression has already been tried with IEZW and SPIHT without motion compensation, the paper for SPIHT merely compare's it to high bitrate MPEG-2... its slightly better but not dramatically so.
"However, at least some semblance of quality American/Western culture is being preserved by the numerous great and recognizeable literary artists we still have alive today"
Ill go along with that.
"Unfortunately such is not the case with modern Japanese culture"
Lets assume thats true for the moment you are going to present the evidence shortly, here it comes...
"For example, the only recent (in the late 20th century) great Japanese cultural celebrity who Americans have been mass exposed to is Akiro Kurasawa. Sadly, commercialized anime is being mass releaed dominating the exposure of true Japanese art and literature. Soon it seems as if true Japanese literature will become extinct as we are losing our connection to their culture. We should be preserving the true art of Japan, not killing it by buying up cheesy anime."
Oh wait I was mistaken, no proof present... all you have shown that the American perception of Japanese culture (according to your standards of course) is in decline. How US centric can you get? You arent being exposed to the creme the la creme of Japanese culture so it might as well not exist????
The Japanese are perfectly able of preserving those parts of their own culture they wish just like you are. Not too many Europeans read any of those great contemporary literary artists of your's either but we sure do mass consume your more popular entertainment. There is little difference between our relationship towards the US than your's with Japan apart from the fact that a little more of the artsy fartsy stuff can get accross due to the absense of the huge language barrier and because the large amounts of popular entertainment and common cultural background has given us a frame of reference, I doubt its because of their higher inherent quality.
How much story do you expect in an hour or so long film? Its just meant as entertainment, not art. Two of those stories are just meant as a small part of their bigger manga Universe anyway. And what the hell is wrong with titilation for its own sake? Are you a prude?
I am not American, so I wonder... That Albees and Updikes you mention who are saving you from cultural death, what percentage of Americans do you think which have voluntarily read them? I wouldnt be surprised if they arent "saving" a whole lot of Americans.
Not that people like me really want to be saved... there is a very good reason we dont seek out translated Basho, it would bore us to death. Its just like with Classical music, apart from the more populist stuff I just dont grok it... doesnt do a thing for me, if thats for a lack of education fine so be it. But I can still enjoy a good movie like the Matrix, or a good anime like Mahou Tsukai Tai or a good Sci-Fi novel like Snow-Crash (actually something like "Use of weapons" is more my thing, but I wanted to pick something even someoene has highly cultured as yourself was likely to recognise). Those suit my coarse taste just fine. In the end this is all about taste, you have yours... keep it.
Maybe its time but guess what... you can go back under your bridge cause this guy didnt stand up to tell them anything about that.
And you have the gall to talk about FUD, pfffttt.
Incompatible with what exactly? Not with BSD, thats for sure.
As for GPL everything is incompatible with it, if you believe the zealots anyway, the so called compatible licenses merely allow sublicensing... in themselves they are not compatible, the code has to change license.
Interbase's license will play with any license which isnt viral:
"You may create a Larger Work by combining Covered Code with other code not governed by the terms of this License and distribute the Larger Work as a single product. In such a case, You must make sure the requirements of this License are fulfilled for the Covered Code."
I think the GPL has a questionable wording in the paragraph which makes it viral BTW... "You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program" linking to an unmodified library is not modification in my book.
A shared bus approach with 16 CPU's would not provide enough bandwith... and NUMA just lets programmers use bad code :)
IMO this machine is much too difficult to master for anyone but the tool developers, and since they are already dealing with so much nitty gritty dealing with the segmented memory directly (and more efficiently) wont add that much more work.
Unless you are talking about numbers for a specific benchmark (Linpack for instance) peak for a 1 GHz Athlon would be 4 GFLOP's using 3DNow!. Thats single precision, but then so is what the emotion engine uses.