Intel Core 2 Duo Vs. AMD AM2 197
ThinSkin writes, "ExtremeTech has an extensive performance roundup across the entire line of Intel Core 2 Duo and AMD AM2 CPUs, from the cheap to the ultra-high end. Both companies bring five processors to the table, ranging from $152 to $1,075, with the mid-range CPUs boasting the best in price/performance. From the article: 'It's clear that Intel's Core 2 Duo lineup offers superior performance across the product line when compared with AMD's Athlon 64. In some applications, even a lower-cost Core 2 Duo can outperform some of the higher-end Athlon 64s.'" The ExtremeTech article is spread over 10 ad-laden pages. You can read it all on the printer-friendly page, but you'll miss out on the pretty graphs.
crypto work (Score:3, Interesting)
Seems core2 is closer to Opteron but not quite there.
Tom
Overclocking... (Score:3, Interesting)
http://anandtech.com/cpuchipsets/showdoc.aspx?i=2
The Core 2 Duos are tremendously and easily overclockable. I upped my performance 25% by changing the FSB from 266 to 333. While this sounds like a significant overclock, for the Core 2 Duo it is actually rather conservative. You juse switch to DDR-667 memory. I'm using the stock Intel cooler and my chips are running just fine temperature wise. People who are more ambitious are going for 400+. When you combine the inherent performance and value in the line with the ease of significant overclocking, AMD isn't even in the same ball game anymore.
Re:A consumer win!-how so? We have lost choices (Score:1, Interesting)
Re:A consumer win! (Score:5, Interesting)
Actually when you calculate performance per dollar, it is closer than most think right now. This article is comparing a $200 Intel processor to a $150 AMD processor. When you compare the $200 AMD to the $200 Intel, not only are they neck-and-neck, but in certain benchmarks, the AMD comes out on top.
Imagine that.
Perhaps those that read articles and think for themselves will see such things. Those that only read headlines and troll won't.
Intel does have a very good processor line on their hands with the Core Duo 2. Even the AMD fans admit that. No one has said otherwise. It is the Intel fans who refused to acknowledge how far they were behind for 4 years. Now both are striving to be the top-dog. AMD claims they will be the best with the 4x4 line soon, and no doubt Intel will respond with a new line of their own.
Meanwhile performance is going up considerably, and prices down at the same time. I built my AMD 3000 system two years ago, and I can't believe what you can build now for the same price.
Re:No 64-bit benchmarks (Score:3, Interesting)
Intel is very open source friendly too (Score:5, Interesting)
However, another interesting thing is that Intel is very open source friendly. Intel's new top of the line graphics adapters (found on some core2duo motherboards) have _FULLY_ open source Linux drivers! That is a _BIG_ thing! You can find more information HERE [intellinuxgraphics.org]. Imagine! Now you can have fully open source OS without any binary drivers messing up your system. These on board graphics adapters are also very fast and capable, so it's a big thing to many of us.
While we're squashing people (Score:1, Interesting)
1) Many of us were using the later betas of Windows 2000 64-bit for the XP platform (MSDN subscribers had access) until Compaq cancelled their agreement with MS
2) Many of us have run Windows XP or Windows servers editions on Itanium
3) Many of us have played XP 360 running a 64-bit Windows 2000 derivative for PowerPC
4) Many of us have been in the Solaris world running 64-bit for a REALLY long time
5) Many of us have used IRIX workstations running in 64-bit for a long time
I will toss you a bone and let you feel a little better for getting the encryption bit correct... but wait a second... in a multitasking environment, the simple cost of task switching AMD64 is more expensive... BECAUSE of the additional registers. You may not realized, but for the most part, almost noone actually uses the standard task state change mechanism of Intel processors favoring manual backup and restore with a far jump instead since TSS related calls are painfully expensive. Now task switching has become far more expensive. Now instead of pushing and popping (not including index and pointer regs) general purpose registers requires pushing and popping 256 bytes compared to the earlier 32 bytes. Of course the faster memory performance will compensate for a bit of it, but let's face it, it still hurts a little.
Now let's get to media encoding... ummm what the hell are you talking about?
You seem to like talking about matrix calculations as if any CODEC actually implements the matrix operations in general purpose registers using the standard ALU.... well some do... strictly for reference implementations, but in reality there are units particularly built for this type of calculation. This is the entire purpose of SIMD, SSE, 64Now!, SSE2, SSE3, SSE4 (and altivec for example elsewhere). The fact is, why in the name of all that is good on this planet would anyone implement a matrix calculation in a execution path locked to 64-bit ALU based instructions when in reality, SSE first of all gives you a minimum of 8 packed registers executing on four 32 bit sets in parallel? As I see from your website, mathematical precision is of importance to you, but in the world of CODECs, there is limited need for anything more precise.
Mathematica, Maple, Mathlab etc... obviously benefit from the new architecture, but come on... are you actually making use of 64-bit integer instructions? If it speeds up anything, it's the UI when running simulations. In reality you're comparing apples to oranges. And not that you've noticed apparently, Mathematica has run on 64-bit for a LONG LONG time.
Ok... you talk about larger virtual address space... that's an excellent topic to discuss... Windows Server has supported PAE for a while, or Physical Address Extension allowing virtual memory to be addressed as 64-bit... this is in fact a documented CPU extension from Intel. Oh.. I'm sorry were you talking about PHYSICAL ADDRESS SPACE? Yes, the Xeon faced certain addressing limitations, but they were circumvented using chipset trickery that allowed PAE to be remapped to larger address spacing externally. This is how companies like Dell, IBM, etc... were able to ship Xeon systems that could address 8Gbytes or more. I don't know the full technical details, but I'd imagine that it is closely related to using fancy paging methods through a more advanced MMU.
Windows Vista will be released in two flavors, one in 32-bit and one in 64-bit. I'm guessing from watching the transition from 16 to 32 bit computing that one thing will be clear, since Microsoft isn't releasing a dual-mode binary (Apple didn't either, their package format just pointed to two executables), that intelligent vendors will rely on either
a) Release an installer that will detect the architecture and install the suitable executable 32 or 64 bit
b) Ignore 64-bit since for the most part,
Re:A consumer win!-how so? We have lost choices (Score:3, Interesting)
And sure, IBM's chips are popular in consoles, but that's mostly because IBM is the only major chip house that will offer to develop custom chip designs. The game console companies help fund the R&D bill and sign-on for some minimum initial shipment, and IBM makes some money. IBM takes existing core logic and pieces it together to optimize performance for consoles, and out pops a new chip design.
But there's a good reason IBM has time for game consoles: IBM doesn't have NEARLY the sales volume of AMD, let alone Intel. AMD doesn't have time for custom chips because they have enough trouble just meeting market demand for their x86 products. They are shooting for 100 million chips next year once their new fab ramps up, and Intel is shooting for 300 million. IBM, by contrast, will sell only 3-20 million chips per-console, per-year, depending on popularity, and those sales rise and fall periodically as consoles get introuced and mature. Even with all three consoles in their pockets, don't expect IBM to sell more then 30 million per-year.
I don't really see what the attraction is to small-market chips like CELL anyway. My prediction on the subject has already come true: way back when IBM announced the CELL, I predicted that by the time it was released, the PC market would have a similar-performance chip on-offer.
With Intel's release of the quad-core Core2 products later this year (Clovertown), my prediction becomes reality. With 2x128-bit SIMD units per core (for a total of 8 128-bit units per chip), Clovertown is capable of the same performance levels as Cell. AMD's K8L, due out next year, is expected to boast similar performance levels.
Re:A consumer win! (Score:3, Interesting)
AMD's 386DX 40 Mhz pushed Intel to release faster 486 chips...otherwise Intel would have ridden their overpriced 486DX 33 forever.
AMD and Cyrix produced Pentium clones which pushed Intel and forced them to reduce prices.
AMD's push to revive Socket 7 (Super 7) with the introduction of the 100 MHz bus and the K6-2 forced Intel to release the Mendocino Celeron. With on-die cache, it was one of the best budget gaming processors ever released. The K6-2's "3D-Now!" instruction set forced Intel to finally admit that MMX was a failure, and release an addentum in 1999 with SSE.
AMD saved us from the hot, expensive, high-latency world that is Rambus. Without AMD pushing the industry-standard PC-133 SDRAM and later DDR SDRAM, Intel would have made Rambus the defacto desktop memory standard.
Fast-forward to today: AMD has been at parity or better for years, and it has slowed their improvements. After the cancelation of the K9 project, AMD sat on its ass, just releasing small improvements for K8. But now, with Conroe wiping the floor, AMD has to get back in gear, and is releasing K8L next year.
It's the way of the business. When you're on-top, you tend to stagnate.