PS3 chip = 1 teraflop

[chibianh]

don't know where to post this... but..

http://news.zdnet.co.uk/story/0,,t269-s2120395,00.html

"Cell's capabilities will allow it to deliver one trillion calculations per second (teraflop) or more of floating-point calculations. It will have the ability to do north of 1 trillion mathematical calculations per second, roughly 100 times more than a single Pentium 4 chip running at 2.5GHz."


that's kinda cool...

[Shrek]

Wouldn't something like that be expensive to mass-produce, which would make the chip more expensive for people to buy?

[peter2002]

Sounds great, but the multi-core technology is very new. It won't come out for 2 years, and will take at least 2 more years before OSs, compilers, and software can take full advantage of the new technology. By then, some new fangled microprocessor design will probably be on the scene.

Peter :)

[gopher]

Originally posted by chibianh
don't know where to post this... but..

http://news.zdnet.co.uk/story/0,,t269-s2120395,00.html

"Cell's capabilities will allow it to deliver one trillion calculations per second (teraflop) or more of floating-point calculations. It will have the ability to do north of 1 trillion mathematical calculations per second, roughly 100 times more than a single Pentium 4 chip running at 2.5GHz."


that's kinda cool...

At last...mathematical proof that the Pentium IV is 50% slower than the G4. The G4 is capable of 15 gigaflops. 100 times slower than a teraflop is 10 gigaflops.

[mrMahann]

lets see it in our macs.

lets see our macs in the same ballpark of crunching computations as an x86 machine.

i know its easier to do work on macs (daily, i use windows and KDE3), but its not gaining us much market share.

[topicolo]

Originally posted by gopher


At last...mathematical proof that the Pentium IV is 50% slower than the G4. The G4 is capable of 15 gigaflops. 100 times slower than a teraflop is 10 gigaflops.

*sigh*:rolleyes: If the P4 is 50% slower than the G4, it would be capable of 7.5gigaflops. 50% of 15 GFLOPS is 7.5GFLOPS. Sheesh.

BTW, if anything, the P4 would be more able to reach its theoretical limit just on the basis of the fact that it has a much faster fsb and ddr/rdram.

[Catfish_Man]

...is purely theoretical. 1 FLOP is a minimum of 4 bytes. 1 Gigaflop is a minimum of 4 billion bytes (3.814.697 Gigabytes). PC133 ram maxes out at roughly 1.05 Gigabytes per second, so a a G4+ can't possibly do anything at 15GFlops unless it's small enough to store in the level 1 cache (maybe the level 2, but I doubt it). A Pentium 4's theoretical limit for non-cached operations is 4 times as fast as a G4+ (4.2GB/sec vs. 1.05GB/sec). The fact that the G4+ is still in the running shows how important cache is.

[Mr. Anderson]

Nice little chip, that's for sure - 16 CPU cores on one chip - but with no indication of clock speed. I'd love to see a multi-core chip for the Mac.....

Any way, if anyone actually read the article, it seems that they acknowledge that the chip design and fabrication would be difficult, but the real test is getting the software to run on it.

While Cell's hardware design might be difficult, it's creating software for the chip that will be the trickiest part of establishing it in the market.

"It's going to take an enormous amount of software development," Doherty said. "We believe the chip architecture is going to be on time and ahead of the software wizardry that is going to really make it get up and dance."


This is futureTech, as far as I'm concerned. It will be a while before we see this, I'd be very surprised if it showed up in 2004.

D

[fragiledreams]

Originally posted by gopher


At last...mathematical proof that the Pentium IV is 50% slower than the G4. The G4 is capable of 15 gigaflops. 100 times slower than a teraflop is 10 gigaflops.

You are a classic (giga-tera-super-ultra) marketing victim, which is not a bad thing because thanks to fools like you my MSc in Marketing has (any) value.

[gopher]

Originally posted by fragiledreams


You are a classic (giga-tera-super-ultra) marketing victim, which is not a bad thing because thanks to fools like you my MSc in Marketing has (any) value.

I'm no fool...anyone with half a brain who knows the gigaflop spec of the G4 can tell that means the G4 is faster. We need a common benchmark. Gigaflop is more reliable than Mhz as the G4 has RISC and Altivec on its side. What good is all that MMX stuff if it can't do simple floating point arithmatic as fast as the G4.

Mhz isn't even reliable as an indicator of speed between Pentiums, let alone Pentiums and G4s.

Floating point calculations though is something that computers are great at. The G4 just happens to be better at it than the Pentium. Give the a G4 a nice floating point calculation it will breeze past the Pentium.

[cc bcc]

IBM, Sony and Toshiba are developing this chip. IBM also said elements of this chip are expected to be in future IBM server chips. Now that's interesting, since IBM chips might end up in future Apple hardware! :D

[Gelfin]

Originally posted by topicolo
BTW, if anything, the P4 would be more able to reach its theoretical limit just on the basis of the fact that it has a much faster fsb and ddr/rdram.

The P4's problems run deeper than the FSB. The chip itself is crippled internally. The fetch/decode/cache architecture can't keep the various execution pipes filled even theoretically, meaning a substantial amount of that fancy silicon is sitting idle at any given moment.

My understanding is that the original design for the P4 did not have this limitation, but that marketing pressures and management deadlines (which have, in recent years, come to dominate over quality engineering at Intel) forced them to cut corners. The P4 as originally designed is said to have been a seriously badassed chip. Too bad they didn't get to make it.

[alex_ant]

Originally posted by Gelfin
The P4's problems run deeper than the FSB. The chip itself is crippled internally. The fetch/decode/cache architecture can't keep the various execution pipes filled even theoretically, meaning a substantial amount of that fancy silicon is sitting idle at any given moment.
It's pretty impressive then that it still manages to piss all over the G4 on most benchmarks, even coming within 70% of the fastest CPUs in the world (Itanium2 and POWER4) in SPEC. :) Efficiency isn't all that important when you've got a chip architecture designed to scale to 10GHz in the next several years.

Alex

[mischief]

Originally posted by alex_ant

It's pretty impressive then that it still manages to piss all over the G4 on most benchmarks, even coming within 70% of the fastest CPUs in the world (Itanium2 and POWER4) in SPEC. :) Efficiency isn't all that important when you've got a chip architecture designed to scale to 10GHz in the next several years.

Alex

I refuse to give two mouse farts of credability to comparissons of Chips not done in a controlled scientific setting on IDENTICAL motherboards. Comparing a dual G4 running on a 133 bus with SDRAM to a P4 Single processor running on AT LEAST a DDR motherboard is just silly. Particularly when AltiVec is very unlikely to be accessed while accelerated proprietary parts of the Pentium probably will. :rolleyes:

Where's an UUbergeek wirehead when you need one?

[alex_ant]

Originally posted by gopher
I'm no fool...anyone with half a brain who knows the gigaflop spec of the G4 can tell that means the G4 is faster. We need a common benchmark.
We have one... it's called SPEC. There are others as well.
Gigaflop is more reliable than Mhz as the G4 has RISC and Altivec on its side.
RISC/CISC distinguisher is irrelevant and meaningless nowadays. "Gigaflop" is a measure of floating point operations / second. It doesn't measure integer ops.
What good is all that MMX stuff if it can't do simple floating point arithmatic as fast as the G4.
CFP2000:

PowerPC G4 1GHz: 178 (peak)
Pentium 4 2.53GHz: 861 (base)

Granted, the PowerPC's results are sure to improve once the test is repeated with Jaguar and GCC 3.x. Even if they tripled, though (which they wouldn't even come close to doing), they will still not hold a candle to the P4.
Mhz isn't even reliable as an indicator of speed between Pentiums, let alone Pentiums and G4s.
No but benchmarks are.
Floating point calculations though is something that computers are great at.
Yes... computers are very good at math.
The G4 just happens to be better at it than the Pentium. Give the a G4 a nice floating point calculation it will breeze past the Pentium.
What I'm sure you meant to say is that it's better at fp than the Pentium for some simple code that has been vectorized explicitly to take advantage of AltiVec and that doesn't demand double-precision results. For double-precision fp, and for relatively complicated fp algorithms that many other chips handle fine, the G4 sucks.

Alex

[alex_ant]

Originally posted by mischief
I refuse to give two mouse farts of credability to comparissons of Chips not done in a controlled scientific setting on IDENTICAL motherboards.
I'm comparing the fastest G4 to the fastest P4. I thought that was what we were on about. If you want to compare using identical motherboards, keep waiting because there will never be identical G4 and P4 motherboards. Shortly after the G4's system bus catches up to the P4's (if ever), the P4's will jump ahead yet again.
Comparing a dual G4 running on a 133 bus with SDRAM to a P4 Single processor running on AT LEAST a DDR motherboard is just silly.
I'm comparing the best of what's available right now. Once the G4 gets a faster system bus I'm sure it will improve. But right here, right now, the P4 destroys the G4, and the P4 isn't standing still either.
Particularly when AltiVec is very unlikely to be accessed while accelerated proprietary parts of the Pentium probably will. :rolleyes:
It's not Intel's fault that their compiler is awesome and PPC compilers suck...

Alex

[alex_ant]

Originally posted by gopher
At last...mathematical proof that the Pentium IV is 50% slower than the G4. The G4 is capable of 15 gigaflops. 100 times slower than a teraflop is 10 gigaflops.
Even if that were not a vast misleading oversimplification, which it is, the number would be 33%, not 50%. 1 - (10 / 15) = 33%.

[dongmin]

Originally posted by alex_ant

Efficiency isn't all that important when you've got a chip architecture designed to scale to 10GHz in the next several years.


Efficiency does matter once you start talking about heat and energy consumption, which may not be an issue with desktops necessarily but definitely is with laptops, sub-laptops, and compact desktops like the new iMac and the Cube.

The issue gets lost in all the discussion regarding cpu clock speeds, but the G3/G4 architecture has allowed Apple a great deal of design flexibility. With P4s, you can kiss goodbye fanless cooling, 4-5 hour battery life in laptops, and cool-a$$ designs like the Cube.

I agree, though, that good design can only go so far. We need faster computers, period.

[alex_ant]

Originally posted by dongmin
Efficiency does matter once you start talking about heat and energy consumption, which may not be an issue with desktops necessarily but definitely is with laptops, sub-laptops, and compact desktops like the new iMac and the Cube.
This is true. Software always seems to increase in bloat proportionately with increases in processor performance, but perhaps in the coming years this trend will slow enough to make putting relatively slow, energy-efficient chips in laptops more of a popular thing to do in the x86 world.

Low power consumption is available in the x86 world, just not from Intel. You can get a VIA C3 which will perform similarly to the iMac's G4, AltiVec aside... you can't get one in such a nice package as the iMac, of course, but that's another discussion. :)

Alex

[topicolo]

Originally posted by mischief


I refuse to give two mouse farts of credability to comparissons of Chips not done in a controlled scientific setting on IDENTICAL motherboards. Comparing a dual G4 running on a 133 bus with SDRAM to a P4 Single processor running on AT LEAST a DDR motherboard is just silly. Particularly when AltiVec is very unlikely to be accessed while accelerated proprietary parts of the Pentium probably will. :rolleyes:

Where's an UUbergeek wirehead when you need one?

Raw processor power performance means about as much as two mouse farts. It won't matter if a G4 is 10x faster than a p4 if it was on a 533Mhz QDR bus with 1066Mhz RDRAM because it isn't available! The main point of these processors is to provide people with a faster way of computing. Thus you have to compare the available platforms to each other. That G4 with kickass ram and fsb will be fast, but it does crap all for me if I can't buy it for any amount of money. I might as well claim that a 4Ghz G5 will be 20x faster than current P4s for the fat lot of good that'll do me

[jwtillema]

*sigh* If the P4 is 50% slower than the G4, it would be capable of 7.5gigaflops. 50% of 15 GFLOPS is 7.5GFLOPS. Sheesh.
Even if that were not a vast misleading oversimplification, which it is, the number would be 33%, not 50%. 1 - (10 / 15) = 33%.
You guys are both wrong. Gopher is the closest to the correct answer, but since you all are arguing technicalities, I'll be anal too.

Read the words carefully.
...he ability to do north of 1 trillion mathematical calculations per second, roughly 100 times more than a single Pentium 4 chip...
Mathematically written:
1000 gigaflops < 100 X + X, X = pentium 4 speed
X > 9.90 gigaflops
At last...mathematical proof that the Pentium IV is 50% slower than the G4. The G4 is capable of 15 gigaflops. 100 times slower than a teraflop is 10 gigaflops.
Mathematically written:
15 gigaflops = 9.90 + 9.90X, X = .515 = 51.5% faster
Another way to look at that (for those who are puzzled) is:
15 gigaflops = 9.90X
X = 1.515 = 1 + .515 = 1 + 51.5% = speed of pentium 4 + magnitude of speed increase

[DannyZR2]

You guys are forgetting how important the L3 cache is. Yes, the P4 can have an increddible 533mhz fsb, but it is still dependant entirely upon that bus to get data, whereas the G4's L3 cache is on a separate direct bus to the chip. (each chip has it's own L3 bus in the case of the Dual 1ghz)

"The L3 cache boosts processor function by providing fast access to data and application code at speeds of up to 4 gigabytes per second (GBps). And the L3 cache is even more effective at speeding up processing because it has a dedicated bus to the G4 chip, making the full 4GBps data throughput always available — unimpeded by bottlenecks caused by other data (unlike Pentium 4-based systems, which don’t have L3 cache — a disadvantage that leads to congestion between various data streams, and slowdowns in the overall rate of data transfer). "

So you have 4GB/s for each of the 2mb caches (per chip).


"The high speed L3 cache with its dedicated bus enables the PowerPC G4 processor to receive data up to five times faster than it could from main memory... so it isn’t idling while waiting for the next task to arrive. "

Everyone is talking about how the G4 is starved for data all the time because of SDR RAM, but you have to remember that the main memory bus isn't the only place data comes from.

"The L3 cache is large enough to store active application code and data. When you run an application, most of the active code for the program — along with most of the data being used — is in L3 cache. "

I'm not sure how small each data chunk usually is, but I don't think many are larger than 2mb. Anyone know about this? Aren't larger chunks broken down most of the time?

I personally will be happy to see DDR memory come to the PowerMac line, if anything, just to satisfy all you morons out there, but I don't think the gains will be as much as many here have been expecting. I'm pretty confident that the G4 isn't as starved for data as so many of you seem to think.

I feel a flame post coming on...

[djwoolf]

This thread was supposed to be about the Cell multi-cored processor. All the processor discusions always turn into P4vsG4 bottom line is that that in not what the discussion is bout.

That said... Think of the possibilities of a multi-core chip. As an avid gamer this could mean better AI for the games, better video quality and mor realistic looking games. Yes it would be awesome to bring such tech to the mac but think about the next $300 PS. Ahhh makes me foam at the moutth just thinking about it.

[Catfish_Man]

Originally posted by DannyZR2
You guys are forgetting how important the L3 cache is. Yes, the P4 can have an increddible 533mhz fsb, but it is still dependant entirely upon that bus to get data, whereas the G4's L3 cache is on a separate direct bus to the chip. (each chip has it's own L3 bus in the case of the Dual 1ghz)

"The L3 cache boosts processor function by providing fast access to data and application code at speeds of up to 4 gigabytes per second (GBps). And the L3 cache is even more effective at speeding up processing because it has a dedicated bus to the G4 chip, making the full 4GBps data throughput always available — unimpeded by bottlenecks caused by other data (unlike Pentium 4-based systems, which don’t have L3 cache — a disadvantage that leads to congestion between various data streams, and slowdowns in the overall rate of data transfer). "

So you have 4GB/s for each of the 2mb caches (per chip).


"The high speed L3 cache with its dedicated bus enables the PowerPC G4 processor to receive data up to five times faster than it could from main memory... so it isn’t idling while waiting for the next task to arrive. "

Everyone is talking about how the G4 is starved for data all the time because of SDR RAM, but you have to remember that the main memory bus isn't the only place data comes from.

"The L3 cache is large enough to store active application code and data. When you run an application, most of the active code for the program — along with most of the data being used — is in L3 cache. "

I'm not sure how small each data chunk usually is, but I don't think many are larger than 2mb. Anyone know about this? Aren't larger chunks broken down most of the time?

I personally will be happy to see DDR memory come to the PowerMac line, if anything, just to satisfy all you morons out there, but I don't think the gains will be as much as many here have been expecting. I'm pretty confident that the G4 isn't as starved for data as so many of you seem to think.

I feel a flame post coming on... ...something like video, or 3d rendering, or gaming relies very little on the cache because it is changing the stuff it's working with all the time. Something like MS Word benefits a lot because the set of things it's working with (text) is relatively small, and completely unchanging. Also, the L3 cache in the G4+ has less bandwidth then the P4's main memory :( (4GB/sec vs. 4.2GB/sec), and the P4 has a larger faster L2 cache. The processor part of the P4 is really not worth a lot (aside from clock frequency) but its memory/cache setup really does kick ass (not quite as much as marketing would lead you to believe, but it is quite nice). The AthlonXP has a really beefy execution core hooked to about average memory (256kb on chip L2, DDR266 ram). The G4+ has a medium processor (weak on DP floating point and clock frequency, pretty good otherwise) with 4 really really kick ass vector units, a very good cache setup (a little low on L2, but it has L3), and really crappy (low end budget PC) ram.

[DannyZR2]

quote:the L3 cache in the G4+ has less bandwidth then the P4's main memory (4GB/sec vs. 4.2GB/sec), and the P4 has a larger faster L2 cache.

The P4 L2 cache is what 512kb? that's nothing compared to the 2mb/chip l3 cache for the g4

[snoopy]

Getting back to the topic of cells, it is likely too early for outsiders to know much. It may not be something that replaces todays processors, but is a means of taking processors to the next level of performance. There may be nothing to prevent cell technology from implementing PPC and AltiVec instructions. With IBMs heavy investment in the PPC architecture, I am sure they would be working on ways to extend the usefulness of the PPC instruction set, not to replace it completely with something totally new.

[alex_ant]

Aww, why do we have to talk about this boring Cell thing? P4 vs. G4 flamewars are so FUN! :D

[topicolo]

Originally posted by jwtillema


Mathematically written:
15 gigaflops = 9.90 + 9.90X, X = .515 = 51.5% faster
Another way to look at that (for those who are puzzled) is:
15 gigaflops = 9.90X
X = 1.515 = 1 + .515 = 1 + 51.5% = speed of pentium 4 + magnitude of speed increase

sure, we made the mistake of the 9.90GFLOPS calculation, but you haven't proven wrong the main point of our posts. All you've stated is that a G4 is 51.5% faster than a P4, not that a P4 is 50% slower than a G4.
15*0.5<9.90
(15-X)/15 = 0.34
where X, the GFLOPS of a P4 = 9.90

therefore the P4 is only 34% slower than a G4

[topicolo]

Originally posted by DannyZR2
quote:the L3 cache in the G4+ has less bandwidth then the P4's main memory (4GB/sec vs. 4.2GB/sec), and the P4 has a larger faster L2 cache.

The P4 L2 cache is what 512kb? that's nothing compared to the 2mb/chip l3 cache for the g4

umm, the P4's cache is running at full processor speed (ie 2.54Ghz), while the current G4, the PPC 7455 has only 256k of L2 cache running at a max of 1Ghz. The L3 cache is only 4ns, meaning that it can only run at a rated speed of 250Mhz (500Mhz effective. ie 1000/4 = 250). More cache means squat if there's no speed to back it up

[mischief]

It's completely obvious from the discussion so far that there is no real way to compare a G4 and a P4 that isn't inherently biassed. In terms of there being no available mobo for the G4 to match the P4's available set.... that just re-enforces my point. You're not actually comparing chips so don't say you are. :rolleyes:

You may as well be argueing about comparing a full-size cargo diesel to a Japanese Motorcycle like a Honda CBR-900. There is just too much difference in what they're designed to do and how power is delivered.

As to the main topic: IBM has been siting on multi-core chip technology and core cross-connect for quite a while and it IS already on the PPC roadmap.

[alex_ant]

Originally posted by mischief
It's completely obvious from the discussion so far that there is no real way to compare a G4 and a P4 that isn't inherently biassed. In terms of there being no available mobo for the G4 to match the P4's available set.... that just re-enforces my point. You're not actually comparing chips so don't say you are. :rolleyes:
You can compare them based on their theoretical maximum performance or on their actual real-world performance. You're apparently trying to compare based on the former, and I'm trying to compare based on the latter. What good is theoretical maximum performance when it's only theoretical and it will realistically never be achieved? Sure a 486 could whip the pants off a P4 if it were overclocked to 5GHz with 500MHz DDR RAM, but it's impossible to do that, so who cares? Same with the G4: Yes, it would probably be as fast or faster than the P4 if it were able to utilize some faster memory and if it were clocked higher, but until it is, who cares.

Alex

[DeusOmnis]

A few years ago Apple predicted a market moving towards 64 bit processing and eventually to 128 bit processing much like our beloved N65's and PS2's.

Unfortunately for Apple, computers are sold based on clock speed, backwards compadibility is a must, and windows dominates the market. Therefore the market stayed at the current 32bit processing.

If Apple correctly predicted the market then the G4 processor would be a faster 32 bit processor or perhaps what the G5 is going to be, a 64 bit processor.

Everyone is talking about how the G5 is going to be super fast, but the truth of the story is that Apple is just going to recess its current technology to gain a marketable quality, clockspeed.

This 128 bit architechture is also the reason why Apple lags in bus speed. It is much harder to create a 400 mhz DDR system bus for a 128 bit architecture than a 32 bit architecture.

[alex_ant]

Originally posted by DeusOmnis
A few years ago Apple predicted a market moving towards 64 bit processing and eventually to 128 bit processing much like our beloved N65's and PS2's.
This is kind of misleading, because you're using different definitions of bitness. 128 bit here refers to the width of the on-chip vector unit. This vector unit still deals with the same 32-bit or 64-bit words as any other 32- or 64-bit CPU; it's just that it processes more than one simultaneously. The Dreamcast, for example, has a CPU that is 32-bit (according to its maximum word length), but it has an on-board 128-bit vector unit which can process four 32-bit words simultaneously (not two 64-bit words or one 128-bit word). Everybody calls the SH3 in the Dreamcast a 128-bit CPU, but by the common definition of the word, it's 32-bit.
Unfortunately for Apple, computers are sold based on clock speed, backwards compadibility is a must, and windows dominates the market. Therefore the market stayed at the current 32bit processing.
It is possible to extend most 32-bit architectures to 64-bit with no loss of backwards compatibility. I know MIPS did this in the early '90s - it is possible to run a 32-bit MIPS binary from 1991 on a modern-day MIPS CPU with no loss of hardware compatibility (software compatibility is another story).
Everyone is talking about how the G5 is going to be super fast, but the truth of the story is that Apple is just going to recess its current technology to gain a marketable quality, clockspeed.
In the short term this is what they'll do. Nobody knows what they'll do in the coming years. If the G5 were to arrive today with the specs it is said to have, it would be super-fast, but it has been vaporware basically forever, so...
This 128 bit architechture is also the reason why Apple lags in bus speed. It is much harder to create a 400 mhz DDR system bus for a 128 bit architecture than a 32 bit architecture.
This is not really true - the problem is that the G4 wasn't designed to support DDR. Presumably Apple has spent all this time grafting a DDR hack onto the new Power Mac motherboards.

Alex

[MacCoaster]

Getting back in the flame war ;)

A lot people who look at the gigaflop numbers and compare to Pentium 4's forget that 15 gigaflops is the theoretical maximum for *TWO* PowerPC G4's. And you're comparing against a single processor Pentium 4. How unfair. Compare to a Dual Xeon, it's no contest; or hell, a dual Athlon MP. Ever wondered why Lucasarts doesn't use G4's for render farms? They use Athlon MPs for render farms; Macs for drawing up and producing stuff. G4 is a great processor, especially for embedded products; but it sure sucks in the desktop market.

Message to Apple: Screw Motorola. Go IBM. Or hell, work with AAI (Apple, AMD, IBM) and you've got a kick ass combination. *sigh* If that would only ever exist. Or just take Power4, add water [actually, Altivec] and other accelerators, produce it in SOI .7micrometers, reduce power intake to around same level as Athlon, and you've got a kick ass PowerPC architecture CPU that can be used in a Mac [to my understanding, since Power4 is just an extremely powerful PowerPC, if I'm wrong, slap me in the face and bitch at me].

[Paolo]

Your wrong... You say your comparing the real world speeds of these computers.
How many fools go and buy a PC for graphic design.... If I recall correctly not many they prefere Macs... word processing doesn't matter as it doesn't require anything... internet doesn't matter as it doesn't require anything... 3d rendering.... okay so this is a function where your arguement may stand, but as far as I know Apple don't go for the 3d rendering market.... that's why PIXAR don't use G4's and everybody knows that!
Okay audio... well processor wise that isn't a huge ask it's just the bits and pieces to allow you to set it up thats the draw back.
and then there's games... well all the games I've played on my mac.. have run just as good if not better on my mac... so that doesn't matter much either.
So when your comparing real world speeds... well you sort of have to think about real world operations... and I find it hard to believe that everybody wants to do full scale rendering on there mac!

So please SHUT UP your posts are sooooo boring!

(as said by DannyZR2: "I feel a flame post coming on")

[Chryx]

Originally posted by gopher
What good is all that MMX stuff if it can't do simple floating point arithmatic

MMX went out of fashion in 1998, I suggest you read up on SSE, 3dnow! and SSE2

all of which DO SIMD on FP data, SSE2 even works with double precision FP


(MMX OTOH was a cheap 'n' nasty hack that didn't help much for pretty much anything.)

ARS Technica writeup on SIMD on the Pentium 4 (http://arstechnica.com/cpu/01q4/p4andg4e2/p4andg4e2-5.html)

[DannyZR2]

Read this from BareFeats.com

http://www.barefeats.com/xserve2.html

"I was curious how the Xserve with its DDR RAM and 2.1GB/s memory throughput would fair running non-server applications. Like, what if I wanted to use the Xserve as a number cruncher or several rack mounted Xserve's as a rendering farm? Is it worth the extra money?

I tested both the dual and single CPU Xserve models. Check it out...

After all the hype touting the Xserve's "phenomenal processing power," I was disappoined in the performance. The more expensive Double Data Rate Memory gives it no measureable advantage. "

As I said before.. I wonder if everyone is putting too much hope in the DDR, when we already have a great architecture with the L3 cache... and if they up that to 4MB per chip.. oh yeah.. that'd be nice...