you'll need quicktime to watch this, but it's a very good presentation about the "mhz myth".
<A HREF="http://www.apple.com/g4/myth" target="_new"> read here </A>
Yes, the PowerPC G4 is an extremely efficient processor but the presentation neglected to mention the register sizes that the CPU can support. The Itanium may be a 10 pipeline CPU but it can accept 64-bits at a time, while the G4 is a 32-bit processor. So the Itanium is probably faster than the G4 in 64-bit apps. If anyone is interested the Athlon has a 10 stage ALU and and 3 15 stage FPU units.
AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor <P ID="edit"><FONT SIZE=-1><EM>Edited by AMD_Man on 10/21/01 09:46 PM.</EM></FONT></P>
I can say one thing, that server is slow as all hell.
Now to the business of things, i watched (95%) of the movie and i got tired. My beef is:
1) no athlon
2) L3 cache shouldn't matter since DDR ram operates at the same frequency as the processors fsb(athlon) therefore it is capable of supplying it with as much data as any l3 cache could. Also i think it would be very unlikely for there to be a throughput bottleneck in the ram unless you were doing something crazy (which the demonstration, image processing, didn't).
3) This is the worst case scenario, I'd like to see the G4 go up against an athlon in ray tracing for instance.
my opinion
<font color=blue>I for one run Quake 3 on a P133(No MMX)</font color=blue>I have no affiliatioin w/ Intel
L3 cache would matter because it would be lower latency than DDR RAM so latency intensive apps would be affected. The 1.4GHz Athlon generally matches the PowerPC G4. I have no link, but that's what I heard on TechTV on "Silicon Spin" hosted by John C. Dvorak, lol.
AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
Hmmmmmmmmmmmmmm Photoshop again. I guesse when I'm playing Quake on a Mac and everybody ask me why, I can say "because it does Photoshop better" And then go on to say "but I never use Photoshop"
Another thing that makes the test unfair is that I believe that the Altivitec optimisations are in use. You know that supposed 128 bit pipeline, I don't claim to understand it. But supposedly it has similar effect as the SSE and so on floating point optimisations.
I thought that the man's lecture didn't actually say anything new, it is just what AMD have been trying to get across.
Another thing, they didn't tell us what the platforms were. I mean they probably mounted the P4 on an i845 chipset or wiht PC 600 RDRAM.
Thirdly, arn't those kind of tests that they are running FPU dependant rather than the core ALU (if this is the case, this explains why they didn't show Itanium which is the processor in FPU performance) and we do know that P4's FPU isn't great, but the ALU (being triggered on the rising and falling edge of the clock) is rather good.
Of course I could be spouting absolute B/S and I am sure you won't have any compunctions about correcting me if I'm wrong.
I wassn't aware that Itanium had a native version of Photoshop. With that said, Intel admits that the X86 portion of the chip (excuse my errors in terminology please) does not perform well; that the P3 or P4 will outperform Itanium in that area. So, why even bring up Itanium as a comparison here?
Mark-
When all else fails, throw your computer out the window!!!
No, you missed my point. What I was trying to say that the test was extremely artificial only using those applications that make the G4 look good (not that apple are alone in this tactic). What I inferred was that they were suggesting that the G4 was supperior to Itanium, Ultra-sparc, P4 and (indirectly) the Athlon. This is clearly not the case firstly the x86 processors do perform better than G4 in other (non-altivitec optimised) applications. Going back to Itanium, which is a very strong performer in FPU code (better than Ultrasparc and MIPS), if they really wanted to do an objective test they would have selected something that could compare the G4's and Itaniums floating point units, although I agree in what you say that the Itanium code would have to be 64-bit and G4's 32 bit. I mentioned it because whatshisname was suggesting (that's what it sounded like to me) that G4 was a superior design. This assertion seemed to be based on an empirical assumption based on G4 having fewer stages and a higher frequency.
Well thats my two pence. I have been subjective then forgive me. Apple annoys me in that it's marketing does (or did) at one point be based on Intel bashing and I have a certain affinity with Intel (an irrational whim really), rather than marketing the product on its own merits.
Charlie
Democracy Bernad, it must be stopped!<P ID="edit"><FONT SIZE=-1><EM>Edited by charliec2uk on 10/22/01 05:12 PM.</EM></FONT></P>
point taken Charlie. To continue in your vein, TechTV did a side by side comparison of G4 vs P4 shortly after Apple's meeting in which they showed the *superiority* over P4. What they noted was that G4 is faster on some, P4 is faster on others, but that considering the mhz of the G4 it is a better performing CPU clock for clock, BUT, as you noted, it really depends on the type of work you plan to do.. even within Photoshop!
Mark-
When all else fails, throw your computer out the window!!!
the point of this post anyways, was not to point out if g4 is better or not, but to show a basic and simple explanation of "more mhz" is better myth. i'm not 100% sure so you can correct me if i'm wrong. but i have not seen amd address the issue head-on like this. i just think it would be a lot better if amd did do it like that, instead of the ratings.
But I think we can both agree that the G4 (and apple systems, barring the iMAC (chuckle), in general) are rather more expensive than the P.C's. So long live AMD, long live Intel, long live Microsoft.... Errm, scratch that last one, got over exited.
But I think we can both agree that the G4 (and apple systems, barring the iMAC (chuckle), in general) are rather more expensive than the P.C's. So long live AMD, long live Intel, long live Microsoft who make our hard earned cash count. [Errm, scratch that last one, got over exited]
You are about to answer a thread that has been inactive for more than 6 months. If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.