From the article: ------------------------------------------------ - "Expect more companies to hop on that bandwagon. Cyrix Corp., which makes low-cost computer processors, says it also is interested in working on Linux, once it finalizes its 3D-Now! technology, a rival version of MMX." ------------------------------------------------ -
My response to the journalist: ------------------------------------------------ - Hi,
I'd like to correct a misconception that you printed regarding 3D-Now! technology and Cyrix.
3D-Now! is a joint venture, spearheaded by AMD corp - and currently supported by their K6-2 and K6-III processors. Cyrix and IDT are also supporters of the technology and will be or already have integrated the technology into their CPUs. The WinChip2 available from IDT already has 3D-Now! support.
3D-Now! was released with the K6-2 in June of 998, and is actually more similar to the recently release SSE (formerly KNI) instructions found in the Pentium-III chip from Intel. Both instruction sets are intended to improve floating point performance - important for 3D-Games, Voice Recognition, Video Compression and other similar applications.
They use a technique known as Single Instruction, Multiple Data (SIMD) to achieve higher performance by allowing instructions issued by the processor to actually affect more than one bit of data simultaneously. In the case of both 3D-Now! and SSE, the improvement is the inclusion of 2 execution units each capable of calculating 2 32-bit floating point values simultaneously - thus, each processor can process 4 floating point instructions at the same time.
Intel's competitors have a rough enough time dealing with Intel's marketing campaign without having to deal with bad press. ------------------------------------------------ -
While I will certainly admit - and this is going over it all one more time for those of you unfamiliar with the architectures of the x86 CPUs (I'm really only familiar with the k6, P2 and M2 cores).
FPU performance:
P2: 2 Execution units, variable latency of 3 to 5 cycles. Per clock cycle peak efficiency: 2 instructions per 3 cycles - 66%.
K6: 1 Execution Unit, low latency of 2 cycles. Per clock cycle peak efficiency: 1 instruction every 2 cycles 50%.
M2: 1 Execution Unit, fixed latency of 4 cycles. Per clock cycle peak efficiency: 1 instruction every 4 cycles 25%.
So, take all three CPUs, and assume everything is equal, and no really tricky tricks are used, and run them at the same speed of 300MHz. That means that you'll get roughly:
198 MFLOPs for a PII core. 150 MFLOPs for a K6 core. 75 MFLOPs for an M2 core.
In actuality the numbers usually aren't far off.
The flip side of the equation is that the Cyrix and AMD chips generally have better integer performance - a holdover in their design plans from 2 years ago when the office application benchmark was king - and office apps use mostly integer instructions. So usually per clock the Cyrix and AMD chips will produce more MIPS than the P-2, but less MFLOPS.
However, coupled with the fact that the P2 core also has some additional features in terms of memory access, and whatnot and the real world bench marks change up somewhat. But not too much really. Especially since the K6-2 with CXT core implements some of those features.
Check out Tom's hardware for some benchmarks of the K6-III and P-III. It's got some great benchmarks of how much SSE instructions will improve the P-II core. However, it's a little light on the 3D-Now! optimized titles for comparison.
Check out Ace's hardware for some K6 core info. It gives you some very real insight into how the cores of the P2 and K6 work, and what effects cache memory has on them.
The one I recall used an HTML form to collect the results over a 24 hour period. I assume that the file of votes would then be read and used as a fitness function against the artwork on display. Then, the top couple of pieces would form the seed of the next population.
In essence, I think they discarded the bottom half of entrants. Then, they used the top half (combining genetic instructions) to generate more, with a chance of mutation.
I recall a very similar project done at CMU via the web in which genetic programming was used to generate art. You could vote on which ones you liked best, and then the higher ranked algorithms were used as seeds for the next set of generated images.
No idea what the URL is anymore, and even if it's still running.
I think a lot of people here are confusing ODBC with SQL. ODBC isn't SQL. ODBC is a layer that allows you to talk with a variety of data sources - SQL or not. You can use it access a variety of data sources - Oracle, Sybase, Informix, SQL-Server, Access, FoxPro, etc. All of which provide for very different protocols, and flavors of SQL.
In the end, it provides you with an environment where you can be less concerned about the protocol and how you've linked to the data source - and less concerned about how you stucture your queries (in that they are not custom coded for it).
Sure, you take a performance hit. But, it works pretty good, and allows you to quickly and easily change the data source on the back end of your application. Makes it nice for prototyping an App using a local db schema defined in Access, and then for roll out actually pointing at a db in Oracle, or SQL server...
And - there are some ODBC drivers that are freely available.
I think they may have changed their web site recently. Because, when I saw the story on TV some weeks ago it seemed much more ominous.
They didn't simply have some "Visualize Maryland Abortionists on Trial!" message. They had a table of abortionists nationwide, with a message extoling (can't recall the exact lingo now) people to take action, and that some already had.
Then, down the table, you'd see that some doctors who perform abortions who had been shot and killed had their names in strike-thru text. Hmm - I wonder what kind of message that is...
Just so you know, better, faster compression generally = more bandwidth. A fast CPU generally enables better, faster compression.
If all communications use some sort of encoding scheme, and the processors that do the encoding are enhanced, doesn't it stand to reason, that systems with better CPUs can concievably make better use of less bandwidth?
Oh hey - waitta second, that's how faster modems work. The digital to analog encoding is more efficient, so I can actually send more over the same line! Oh, and if I compress what I send over that encoding format, I can even send more.
Yeah - the ad campaign sucks. But, using an Intel codec that utilizes MMX sure makes streaming media over a lower bandwidth connection a whole lot better. And - now that they have added KNI, who knows, maybe more improvements are on the way.
So yeah, a faster processor does enhance your web based experience, and can even improve non web based internet experiences.
so Intel embeds an ID into their chip. BFD. There are a whole bunch of other IDs in your system that software can get at in various ways. I merely see this as a way to help restrict the deployment of software. Example:
You call up some vendor and ask them to send you a copy of their software (or via the web) - and they ask for your CPU ID, they then ship you a copy, or email it, and only you can run it.
It's sort of like a PGP key.
Now - if you think that this ID could then somehow be tied to you, and your personal information, that's silly. More PCs are bought by businesses than homes. So - at a business, they'll get a lot of systems based on PIIIs. The system is not yours (even if you're sitting there using it) - it could be moved to someone elses desk tomorrow. Or, someone else could be sitting at your desk tomorrow. This thing in no way ties you to a machine and vis-versa.
I guess it was a bad idea to ever encode your bank acct on that plastic card?
When they were talking about on-line commerce - I really don't think they were talking about something as trivial as you making credit card purchases from some porn site!
What we're talking about is yet another ID that physically allows the decryption, or enables the running of something on that system and that system only.
- Porter
PS: I use an AMD chip because I support the underdog, and I wanted to save some $$. I won't be getting a PIII anytime soon, because I'd have to get a whole new MB, etc.
Slashdot Mirror
From the article:- - - -
- -
- -
3 /index.html
n depth.html
-----------------------------------------------
"Expect more companies to hop on that bandwagon. Cyrix Corp., which makes low-cost computer processors, says it also is interested in working on Linux, once it finalizes its 3D-Now! technology, a rival version of MMX."
-----------------------------------------------
My response to the journalist:
-----------------------------------------------
Hi,
I'd like to correct a misconception that you printed regarding 3D-Now! technology and Cyrix.
3D-Now! is a joint venture, spearheaded by AMD corp - and currently supported by their K6-2 and K6-III processors. Cyrix and IDT are also supporters of the technology and will be or already have integrated the technology into their CPUs. The WinChip2 available from IDT already has 3D-Now! support.
3D-Now! was released with the K6-2 in June of 998, and is actually more similar to the recently release SSE (formerly KNI) instructions found in
the Pentium-III chip from Intel. Both instruction sets are intended to improve floating point performance - important for 3D-Games, Voice
Recognition, Video Compression and other similar applications.
They use a technique known as Single Instruction, Multiple Data (SIMD) to achieve higher performance by allowing instructions issued by the processor to actually affect more than one bit of data simultaneously. In the case of both 3D-Now! and SSE, the improvement is the inclusion of 2 execution units each capable of calculating 2 32-bit floating point values simultaneously - thus, each processor can process 4 floating
point instructions at the same time.
Intel's competitors have a rough enough time dealing with Intel's marketing campaign without having to deal with bad press.
-----------------------------------------------
While I will certainly admit - and this is going over it all one more time for those of you unfamiliar with the architectures of the x86 CPUs (I'm really only familiar with the k6, P2 and M2 cores).
FPU performance:
P2: 2 Execution units, variable latency of 3 to 5 cycles. Per clock cycle peak efficiency: 2 instructions per 3 cycles - 66%.
K6: 1 Execution Unit, low latency of 2 cycles. Per clock cycle peak efficiency: 1 instruction every 2 cycles 50%.
M2: 1 Execution Unit, fixed latency of 4 cycles. Per clock cycle peak efficiency: 1 instruction every 4 cycles 25%.
So, take all three CPUs, and assume everything is equal, and no really tricky tricks are used, and run them at the same speed of 300MHz. That means that you'll get roughly:
198 MFLOPs for a PII core.
150 MFLOPs for a K6 core.
75 MFLOPs for an M2 core.
In actuality the numbers usually aren't far off.
The flip side of the equation is that the Cyrix and AMD chips generally have better integer performance - a holdover in their design plans from 2 years ago when the office application benchmark was king - and office apps use mostly integer instructions. So usually per clock the Cyrix and AMD chips will produce more MIPS than the P-2, but less MFLOPS.
However, coupled with the fact that the P2 core also has some additional features in terms of memory access, and whatnot and the real world bench marks change up somewhat. But not too much really. Especially since the K6-2 with CXT core implements some of those features.
Check out Tom's hardware for some benchmarks of the K6-III and P-III. It's got some great benchmarks of how much SSE instructions will improve the P-II core. However, it's a little light on the 3D-Now! optimized titles for comparison.
http://www.tomshardware.com/releases/99q1/99022
Check out Ace's hardware for some K6 core info. It gives you some very real insight into how the cores of the P2 and K6 work, and what effects cache memory has on them.
http://www.aceshardware.com/articles/x86/k6-3_i
- Porter
The one I recall used an HTML form to collect the results over a 24 hour period. I assume that the file of votes would then be read and used as a fitness function against the artwork on display. Then, the top couple of pieces would form the seed of the next population.
In essence, I think they discarded the bottom half of entrants. Then, they used the top half (combining genetic instructions) to generate more, with a chance of mutation.
It wasn't real time, and certainly predated Java.
- Porter
Hey -
I recall a very similar project done at CMU via the web in which genetic programming was used to generate art. You could vote on which ones you liked best, and then the higher ranked algorithms were used as seeds for the next set of generated images.
No idea what the URL is anymore, and even if it's still running.
- Porter
ODBC -
I think a lot of people here are confusing ODBC with SQL. ODBC isn't SQL. ODBC is a layer that allows you to talk with a variety of data sources - SQL or not. You can use it access a variety of data sources - Oracle, Sybase, Informix, SQL-Server, Access, FoxPro, etc. All of which provide for very different protocols, and flavors of SQL.
In the end, it provides you with an environment where you can be less concerned about the protocol and how you've linked to the data source - and less concerned about how you stucture your queries (in that they are not custom coded for it).
Sure, you take a performance hit. But, it works pretty good, and allows you to quickly and easily change the data source on the back end of your application. Makes it nice for prototyping an App using a local db schema defined in Access, and then for roll out actually pointing at a db in Oracle, or SQL server...
And - there are some ODBC drivers that are freely available.
- Porter
I think they may have changed their web site recently. Because, when I saw the story on TV some weeks ago it seemed much more ominous.
They didn't simply have some "Visualize Maryland Abortionists on Trial!" message. They had a table of abortionists nationwide, with a message extoling (can't recall the exact lingo now) people to take action, and that some already had.
Then, down the table, you'd see that some doctors who perform abortions who had been shot and killed had their names in strike-thru text. Hmm - I wonder what kind of message that is...
Hey,
Just so you know, better, faster compression generally = more bandwidth. A fast CPU generally enables better, faster compression.
If all communications use some sort of encoding scheme, and the processors that do the encoding are enhanced, doesn't it stand to reason, that systems with better CPUs can concievably make better use of less bandwidth?
Oh hey - waitta second, that's how faster modems work. The digital to analog encoding is more efficient, so I can actually send more over the same line! Oh, and if I compress what I send over that encoding format, I can even send more.
Yeah - the ad campaign sucks. But, using an Intel codec that utilizes MMX sure makes streaming media over a lower bandwidth connection a whole lot better. And - now that they have added KNI, who knows, maybe more improvements are on the way.
So yeah, a faster processor does enhance your web based experience, and can even improve non web based internet experiences.
- Porter Woodward
OK -
so Intel embeds an ID into their chip. BFD. There are a whole bunch of other IDs in your system that software can get at in various ways. I merely see this as a way to help restrict the deployment of software. Example:
You call up some vendor and ask them to send you a copy of their software (or via the web) - and they ask for your CPU ID, they then ship you a copy, or email it, and only you can run it.
It's sort of like a PGP key.
Now - if you think that this ID could then somehow be tied to you, and your personal information, that's silly. More PCs are bought by businesses than homes. So - at a business, they'll get a lot of systems based on PIIIs. The system is not yours (even if you're sitting there using it) - it could be moved to someone elses desk tomorrow. Or, someone else could be sitting at your desk tomorrow. This thing in no way ties you to a machine and vis-versa.
I guess it was a bad idea to ever encode your bank acct on that plastic card?
When they were talking about on-line commerce - I really don't think they were talking about something as trivial as you making credit card purchases from some porn site!
What we're talking about is yet another ID that physically allows the decryption, or enables the running of something on that system and that system only.
- Porter
PS: I use an AMD chip because I support the underdog, and I wanted to save some $$. I won't be getting a PIII anytime soon, because I'd have to get a whole new MB, etc.