Slashdot Mirror
AMD Roadmap for Coming Year and Beyond
nexex writes: "With a new year comes new products, and AMD certainly has some new toys for us to drool over. The first of 2002 will see the release of "Thoroughbred," a version of the Athlon XP chip made on the more advanced 130-nanometer manufacturing process. The chip will cover 80 square millimeters in area, or 65 percent of the space of the "Northwood" Pentium 4 coming from Intel in early January. That chip measures 116 square millimeters, according to AMD estimates.
For more, including info on Clawhammer, Sledgehammer, and all the Intel bashing you can handle, see here." I hope they don't really mean that "these new chips will also consume less heat than current AMD notebooks chips."
"My biggest fear is that Intel will come out with a 32-bit processor with 64-bit extensions because it is the right thing to do," Sanders said. "The Itanium it turns out is a niche product...We are going to have a role in the industry because we better fulfill Microsoft's needs."
the Itanium is a niche product now. in a few years i expect its time will come. 64-bit is not cool now but eventually OEMs are going to lean that way for upward compatibility. remember that the PowerPC existed in relative obscurity for a while too, and now it's the basis for what are probably the best UNIX machines on the market.
I've bought only AMD processors for years now (starting with a K6-200), and I've never had any problems with the systems in which they were used. It's the result of not getting the absolute cheapest motherboards and other components for these systems. I've seen plenty of Intel-based systems crash and burn, but they were usually dollar-engineered boxen with shitty motherboards (usually PC Chips and similar, though I've had a few MSI boards go south as well).
(I could make some wisecrack about the FDIV bug or the 820 MTH SDRAM compatibility debacle, but I won't. :-) )
20 January 2017: the End of an Error.
Maybe AMD has a new angle on power consumption. Maybe their proccessors extract thermal energy from the surrounding atmosphere to power the chip.
Or maybe not.
That way, the underlynig hardware architecture can be changed at will with little or no impact on OSes or apps. I think that it was a mistake for Itanium to expose strange hardware features to the software compilers. It's too inflexible.
AMD is not doing fine. They had a net loss of two hundred million dollars last quarter. Furthermore, they admitted that they lost about 1% marketshare to Intel in that quarter. My predicition is that they will exit the microprocessor business if the PC industry doesn't pick up within 4-5 quarters. Their cost structure just doesn't support it.
I wish that I could believe this. I've had some difficulties with ALL my recent intel based machines. My lab had 4 Abit BP6 motherboards with the well documented capacitor problems. These motherboards proved unstable when installing different flavors of Linux, each with its own threshold of instability. Then I had one machine with an Asus P2B-DS (probably one of THE MOST STABLE MOTHERBOARDS BY REPUTATION) which had one of the PIII 500's go south after 3 years and lost my data (boy did that suck, thanks Intel!).
I got a cut rate 1 GHz Athlon system this summer and it strongly outperforms all the pentium systems (I attribute it to the DDR memory, but the CPU is blazingly fast compared to Celeron 500s and PIII 500s of 2-3 years ago).
The EPIC instruction set architecture of Itanium/McKinley is not a good match for Java Virtual Machines....at least thats what I read in a technical article about IBM's Power4 architecture. Apparently JVM's can't take advantage of VLIW as well as compiled code can, and this makes sense because Java is compiled to machine-code on the fly. Like it or not, Java is a major player in today's software technology. If AMD continues to excel with IA-32 (which is a decent match for Java), it will help Java as well as AMD...
Simple answer as to why they don't use Alpha or StrongARM, Intel doesn't like non-Intel instructions sets. They have some good reasons for this, ie it keeps control over the instruction set in-house and prevents them from getting screwed over by some other company. I suspect that there may also be a bit of political/marketing type stuff going on as well, in that having their own instruction set means that they have much more control over who produces competing products.
In any case, StrongARM will continue to live on, under the X-Scale name, and Intel does have some rather impressive plans for these chips. Alpha will be put to rest. A bit of the technology might be incorporated into the Itanium line, but probably not much since the two designs are quite different. The real thing that Intel gained through the whole Alpha deal with Compaq was that they acquired some of the brightest CPU designers around and some very good compiler writters (the latter being VERY important for the Itanium).
If you dig deep into Tom's Hardware Guide, you'll also find that Tom is a nutcase that knows VERY LITTLE about computers. He also has a VERY short temper and tends to paint in a good light whatever company is giving him the red-carpet treatment for the month. Tom runs his site for ne purpose, to make a LOT of money, and he does VERY well at that. He probably has one of the most profitable non-pr0n sites on the net.
Anyway, as far as stability goes, VIA, SiS and ALi chipsets have always had more ups and downs then Intel chipsets. Intel hasn't been perfect either, both the i810 and i820 had horrible problems when they first came out (and the i820 never really did go anywhere), and even the old 440LX chipset had it's share of problems. However by and large Intel has been fairly consistant with their chipsets, while the Taiwanese guys have been a bit more over the map.
Another MAJOR issue when it comes to stability of a system is market share. Intel traditionally had the most market share when it came to chipsets, so the third party sound card, video card, NIC, etc. manufacturers test their products against Intel chipsets first and foremost. Testing against VIA, ALi and SiS chipsets used to be a secondory objective at best. Now, I know that some people will jump on this and say that if VIA doesn't work exactly like Intel it's VIA's fault, but really that ain't always so. As the old saying goes, "Standards aren't". VIA and Intel could follow the PCI spec exactly to the word 100% of the way and be TOTALLY incompatible. Actually that's just what happened when PCI first came out, though now things are much better. Still, there are a lot of cards out there that have buggy drivers which only work due to some quirk in some chipsets, and when paired with a different chipset that doesn't have said quirk, things go wrong. Case-in-point, the "VIA" data corruption bug that was caused by buggy Sound Blaster Live! drivers.
The smaller the design, the lower the voltage you need, and the less power you consume.
However, manufacturers also take advantage of the shrink to up the frequency, bringing up the power consumption. We're still a ways off from having a thermal crisis. AMD still hasn't started using "thermal spreaders", and they're doing fine. Once they put the thermal spreaders on, they will be able to dissipate heat more quickly.
For those that think we're to the end of the road for air-cooled processors, no, there are heat sinks/fan combos today that are much more powerful than a chip needs without overclocking - and there are still many improvements to be made. For example, we could use larger fans at lower RPMs to move lots and lots of air without much noise at all - how does 60 CFM at 32 dB sound? (it sounds pretty quiet.)
Also, there are other ways of getting more benefit from air-cooled heat sinks. Most heat sinks do well with the addition of a copper plate on the bottom, for reasons too lengthy to go into here. And, by using well-designed shrouds, you can up the effectiveness even more.
So, why don't they make heat sinks like that today? The same reason they didn't make modern heat sinks 5 years ago - they didn't need them. With a small amount of thought and engineering, I believe that we can at least double the wattage of a chip without too many problems.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.