Slashdot Mirror

Right now, you will not see much, if any, difference in performance between AGP 4X and AGP 8X...

"My 4 THz Intel Pentium IIIVIXXX is father then your 16 KHz G101"

For those of you who have not read ALL of the CPU articles at ArsTechnica. Go there now and do so. Before posting any of your inane babble about clock speed and processor power.

It IS true that Motorola has fallen behind Intel - sort of.

There are other advantages to hardware other then Intel based systems.

Since this is an Apple thread I'll focus there - One of the most note worthy (My opinion) Is apple's System controller.

Go READ the articles at ArsTechnica!

Rather than re-writing I'll simply cut & paste.

The PCI bus is what really impressed me.

Direct PCI bus: In another example of superior architecture, the Power Mac G4 optimizes PCI performance by connecting the PCI bus directly to the system controller. In a typical PC architecture, PCI devices connect to the I/O controller through a bridge, a bottleneck in the data path where all connected PCI devices are slowed down to avoid overloading the system controller. Going through this bridge constrains PCI throughput to 133Mbps (the bus speed on Pentium 4 systems), even with otherwise fast PCI devices. This slowdown of data to and from PCI devices results in greater overall system latency. The Power Mac G4, on the other hand, features a direct 266-MBps bus to the PCI slots to guarantee high throughput and low congestion ? in effect, lowering latency. The Power Mac G4 also supports write combining, which allows write instructions to be grouped into one large instruction, further increasing data throughput.

Then Apple oficially slams PC architecture.

On the Power Mac G4, FireWire, Gigabit Ethernet and even the ATA/100 bus are built into the system and integrated directly into the system controller. (The ATA/66 bus has its own controller.) This dedicated connection reduces PCI congestion and guarantees low latency, resulting in optimal FireWire, Ethernet and hard drive performance. And as a side benefit, it also keeps the computer?s PCI slots free for your specialized audio and video cards instead of using them to provide basic technologies.

I got this info here.

Go READ the articles at ArsTechnica!

Apple is not the end all - be all of systems. Two of the greated systems are made by DEC & H/P. The UltraSparc kicks the crap out of anything Motorola & Intel have to offer.

And let's not forget the Alpha. The Pentium - Pentium III architectures were based on technology stolen from DEC. Technology that Intel is still paying for today.

It basically falls down to system preference. Mac users DO NOT CARE if you can build a PC for $400. Mac users DO NOT CARE if only a few of the best selling game titles are ported to the system.

Having more game titles available is a Good Thing - naturally -but I find myself being... PRODUCTIVE instead of having my time eaten away by games - Linux users also what I'm talking about - unless they've downloaded BZFlag or Crack Attack.

Go READ the articles at ArsTechnica!

According to this paper, the TCPA chip is on the LPC bus and is accessed by I/O mapped registers. LPC is the "Low Pin Count" bus, a 4-wire bus designed as a cheap way to link the CPU to low-bandwidth stuff like keyboards, serial, parallel, floppy, and maybe USB ports. It maxes out around 5 MBytes/sec -- nowhere near enough to keep up with a hard disk or a 100Mbit link. See the LPC bus spec if you want the gory details.

one of the main problems would be the timing for the system and the heat!!!!!!! 4 PIII cores running 2x2 would not only generate somewhere around 200 watts of heat, and would require a 300 to 400 Watt psu for a bare minimum configuration

Don't take this as a disagreement, I do indeed think heat dissipation would (and already does) make one of the biggest problems. However, according to the P-III 1.13-1.40 spec sheet, the .13 micron P-III/1400 only dissipates 32.2W, for a total of 128.8W over four cores. That includes the cache as well.

For comparison, the Athlon XP 2600+ gives off 68.3W, the P4 3.06 sucks 82W, and Intel's next Itanium, the Madison, will nicely heat your computer room at a whopping 130W.


My opinion: Itanium does the job, and if people would spend the time it would take to learn a new archecture, it would be a nice, fast chip to start from.

I agree with you that, technologically, the Itanium looks rather impressive. However, even taking into consideration that it (well, at least the upcoming Madison core) does 6ops/clock compared to the P4's 2ops/clock, that still leaves it short for raw power at only 1.5Ghz (since, by the time Intel starts shipping in quantity, the P4 will certainly have passed 4.5Ghz). I understand that clock speed doesn't mean everything, but clock speed times throughput per clock *does* give a pretty good indication of its upper limit.


somebody give me theirs.. I'm broke again :(

Okay... A quick tip for getting properly-mounted Athlon heat sinks, without risking damage to the chip... Buy a motherboard combo. I suppose this doesn't apply if you just upgrade an existing CPU, but if you need a new motherboard, get them at the same time, from the same place. That way, not only will you not risk a crushed chip, but if it dies of heat within a few days, just send it back and get a free replacement. :-)

I don't think many apps are using this yet. Intel has a FAQ on this technology here, and they even have a Windows driver to support this.

I don't think many apps are using this yet. Intel has a FAQ on this technology here, and they even have a Windows driver to support this.

IBM
HP
Intel,
Crystal PC,
Force Computer,
RadiSYS

There are many many Intel server vendors but there
are only handful who makes NEBS compliant ones.

How does Intel's Hyperthreading Technology differ from the dual core? I realize the obvious, such as one in in the Pentium line and the other in the Itanium, and the physical differences of packaging.

But how how different will the architecture of a dual-die chip differ from hyperthreading, such as which CPU components will be shared (like cache, instruction decode/scheduler, etc.)?

Also would the Linux kernel's logical processor abstraction used to enable hyperthreading support (see IBM developerWorks Article) also continue to work effectively with a dual-die chip?

Also, according to Intel here:

AGP 8x technology is intended to be the last parallel interface step that meets the industry's requirements before transitioning to a PCI Express-based serial graphics solution in 2004. The PCI Express architecture is a high-speed, general-purpose, serial I/O interconnect that provides a unifying standard, consolidating a number of I/O interconnects within a platform.

Maybe this is a feature of newer releases of gcc, but I've never heard of -j doing auto SMP. There is a -j option for parallel makes with gnu make, but this is only for the compilation and not runtime.

The portland group compiler and the intel compiler. Do support some auto-parallalization via openmp and threads.

My OS course is getting a bit fuzzy already, but we were lucky enough to have a visiting professor from a real university teach it. Profesor Tsuni was one of the best teachers I had.

But back to the point, and excuse the "obviousness" of the questions. But HT sounded like a way to more efficiently use the pipelines on modern processors by allowing multiple threads to work on them.

And here is the fuzzy part, or maybe I'm just not remembering correctly. Do the multiple threads need to be in the same process? If so, I remember the linux kernel threading actually throws threads out as new full processes, and I'm unsure of how the CPU can track that. Or is the scheduler smart enough to send processes down the queue in an order where threads that can share processor time easier are sent together?

Also, if some of the posts are correct it seems that multiple processors show up in Top. Off hand I wonder if this hamper or help OpenMosix's algorithms that decide where to place processes to run.

Here is the associated press release from Intel about the HT in 3 Ghz P4's. I have seen screenshots of Windows task manager showing (2) CPU performance graphs.

I made the switch when doing work for my Ph.D. thesis (nuclear physics).

Here is the link: Intel.com

where are the new 486dx's for me to build cheap routers out of?

Right here.

Is there faster ethernet available that I've not heard about?

Well, there's 10 gigabit Ethernet, and Intel are now sampling a card that supports it.

Though now it seems AMD is taking it easy for awhile, so that benefit may have been short-lived.

Hell yeah I wish I could have such a relaxed time, hanging around all day doing litle more the designing an entire new x86 chip with new instructions and a totall new platform design, getting test silicon out, squashing bugs in that, working with vendors like via which couldn`t survive if it wasn`t for you being easy on CPU bus IP issue compared to others and other chipset designers which need loads of help becouse they are new to the chipset arena or just wanna scale your platform way beond what you was thinking of when you started your "lets add 32 bits, how hard can that be" adventure while you are short on cash.

This just when your main competitor has just done the imposible (rather then updating the same boring pentium2 design, design a chip that makes no sense whatsoever except for one thing, upping clock speeds just as fast as consumers can afford and marketing can market).

And all of this while you try to inovate on "normall chips" to (pxa and memory) that all of a sudden noone wants anymore becouse just while you where bussy negotiating you way trough cross licensing agreements with intel then there is an "economic downturn" (hired all these bright people that think up intergrated memory controlers and stuff but none of them can think of such a simple thing).

Then there is the small issue of having to get rid of 10-20% of your staff just after you milion dollar production line was at risk of getting washed away with the rest of dresden, but overall quite relaxed.

The article linked to above links in turn to a much better page on Intel's site. It's practically a howto on using this case to build a kick-ass (and quiet) multimedia pc for the living room.

Itanium exists.

It shows that none of them are really suitable for a probe launched in 1972. The 4004 was only introduced in 1971, and the 8008 in 1972. The 8080 came in 1974, and the 8086 and 8088 in 1978.

is a chart of all the major families, but it doesn't go into so much detail as the Intel link.

Of course, computer manufacturers could never achieve that sort of reliability. Pioneer 10's systems were actually designed and built by highly trained monkeys.

Mobile athlon xp

Mobile pentium 4

-Kufat

I'm having a hard time trying to figure out why you are relating the AMD/Intel speed battle with anything that has to with Moore's Law. Here are a couple of resources for you to read up on:
http://www.intel.com/research/silicon/mooreslaw.ht m
http://www.intel.com/research/silicon/moorespaper. pdf

Moore's law has nothing to do with speed. All it states is that the number of transistors on a manufactured IC will double every couple of years. Note that chip speed is never mentioned.

Also, if you think that Intel just reached the 3GHz level using an unoptimized design, you are most certainly wrong.

I'm having a hard time trying to figure out why you are relating the AMD/Intel speed battle with anything that has to with Moore's Law. Here are a couple of resources for you to read up on:
http://www.intel.com/research/silicon/mooreslaw.ht m
http://www.intel.com/research/silicon/moorespaper. pdf

Moore's law has nothing to do with speed. All it states is that the number of transistors on a manufactured IC will double every couple of years. Note that chip speed is never mentioned.

Also, if you think that Intel just reached the 3GHz level using an unoptimized design, you are most certainly wrong.

Andy Grove, Intel Chairman
Gordon Moore, Intel Chairman Emeritus
Gordon Moore on his law, in which he boldly predicts it will hold another score years.

Andy Grove, Intel Chairman
Gordon Moore, Intel Chairman Emeritus
Gordon Moore on his law, in which he boldly predicts it will hold another score years.

Andy Grove, Intel Chairman
Gordon Moore, Intel Chairman Emeritus
Gordon Moore on his law, in which he boldly predicts it will hold another score years.

Richard Feynman's address to the American Physical Society is a good intro to the physical limitations of miniaturization as it applies to Moore's Law. Also intersting, is the Law from the Horse's mouth found on this Intel page.

What I was told by the HP folks was different. They said that very little of the existing Intannium[sic] I will be evident in the new chip set.

And there's probably not much (if anything) left of the P6 core (Pentium Pro, Pentium II, Pentium III, many Celerons) in the Pentium IV, and not much left of the P5 core (Pentium, Pentium MMX) in the P6 core - but they all implement the x86^H^H^HIA-32 instruction set, albeit with various additions over time (MMX, SSE, and assorted other stuff such as conditional moves and a 64-bit compare and exchange).

The Itanium 2 implementation of the IA-64 instruction set might not share much with the original Itanium implementation IA-64 instruction set, but that doesn't mean that they don't implement the same core instruction set.

In fact, the Intel FAQ on Itanium 2 explicitly says:

Q2. How is the Itanium 2 processor different than the original Itanium processor? What kind of performance can I expect?

A2. The Itanium 2 processor builds on the Itanium architecture with large integrated on-die cache and additional execution units while maintaining compatibility with Itanium-based software.

(emphasis mine).

The incremental cost of electricity for a computer that is idle verses one running at 100% CPU is actually very, very small.

If we assume that the CPU draws 60W more at 100% use than at 0% (Intel lists maximum heat disipation of 60W for the P4), then 8,000 computers would consume a total of 480kW. Sounds like a lot, right? Now consider that so far today, California has had a maximum power draw of 28,000GW, which is 58 *million* times more than 480kW. And that's just one state.

Of course there is, you insensitive clod. As a matter of fact, i ran one for over a year. Yeah, it's just a 166 marked as a 150, but officially it was a 150! http://developer.intel.com/design/pentium/datashts /241997.htm - Intel spec site for pentium processors

That has to be wrong.

Back in 1986, Intel and Sandia built a 1 terraflop computer, capable of 1 trilion (1,000,000,000,000) operations per second. Aristotle died around 322 B.C.E. or 2324 years ago.

Under one interpretation of Moore's law, the number of operations per second doubles every 3 years or so. Working backwards, that means Aristotle's computer was capable of one operation every 10^213 years. The first computer capable of one operation per second would have had to have been built around 1882.

Conclusion: Aristotle's work must have been all theoretical.

Here's the XScale link and the Solar PC link.

Intel's XScale site is here: http://www.intel.com/design/intelxscale/

Well not one, but.. I know this isn't a big name chip, at least in the PC world, but it's sad to see chip makers go away. One day AMD says it's backing off in its fight, now this one. Let's hope somebody else comes up soon and fights Intel While it's mentioned, anyone know of any up-and-coming chip makers?

Between that and some of the other displays it was worth the trip

Between that and some of the other displays it was worth the trip

In fact, probably even the new EPIA-M board is a better deal for many applications; the EPIA-M costs $160 with processor, uses a 933MHz C3 (Pentium compatible), is tiny, and uses comparatively little power. And if you buy one of those, you don't even give money to the other monopoly.

Wrong. You are such a FUD/conjecture machine it's unreal.

The Intel shrinks of the P3 (.13) with the integrated heat spreader are quite cool to the touch under load. In fact, the FIRST CPU Intel shrunk was the Pentium III. You know, you can READ about CPUs. http://www.intel.com/intel/finance/pricelist/ , http://developer.intel.com/design/PentiumIII/specu pdt/ and for P4 lovers, the specification updates for the p4, http://developer.intel.com/design/Pentium4/specupd t/.

Now there is similar documentation available from AMD as well. Now instead of getting anecdotal evidence, (Which can show you clearly there are many holes in this guys Pentium 3 theory), you can see steppings, dissipations, packaging and thermal characteristics for CPUs, not some asshole's rendition of fact into half truths. You want to talk about heat, talk about dissipation first.

Intel could put out a 1.6GHz Pentium III tomorrow [The 1.4 is able to be clocked at that frequency]. They stopped the P3 at 1400 with 512cache for a reason - it started to compete with Pentium 4.

Do you expect anyone to believe you have ever read EETimes or have a sub to Microprocessor Report? I mean, you are poorly summarizing gamer sites here continually.

By the way, for anyone thinking of this guys trash, the Athlon shrinks and the hammer chips will run cooler. Maybe it wont be up to snuff with Intel.

Now the biggest problem with your filth is that Hammer specs aren't out. system integrators need thermal information to make designs. And when Dell and friends get the thermal data, along with everyone else, then you can talk about what kind of heat output AMD-64 will have.

Everything you say is conjecture, crap, fluff. You don't work for AMD, and judging from your intellect, if you did, they laid you off.

Wrong. You are such a FUD/conjecture machine it's unreal.

The Intel shrinks of the P3 (.13) with the integrated heat spreader are quite cool to the touch under load. In fact, the FIRST CPU Intel shrunk was the Pentium III. You know, you can READ about CPUs. http://www.intel.com/intel/finance/pricelist/ , http://developer.intel.com/design/PentiumIII/specu pdt/ and for P4 lovers, the specification updates for the p4, http://developer.intel.com/design/Pentium4/specupd t/.

Now there is similar documentation available from AMD as well. Now instead of getting anecdotal evidence, (Which can show you clearly there are many holes in this guys Pentium 3 theory), you can see steppings, dissipations, packaging and thermal characteristics for CPUs, not some asshole's rendition of fact into half truths. You want to talk about heat, talk about dissipation first.

Intel could put out a 1.6GHz Pentium III tomorrow [The 1.4 is able to be clocked at that frequency]. They stopped the P3 at 1400 with 512cache for a reason - it started to compete with Pentium 4.

Do you expect anyone to believe you have ever read EETimes or have a sub to Microprocessor Report? I mean, you are poorly summarizing gamer sites here continually.

By the way, for anyone thinking of this guys trash, the Athlon shrinks and the hammer chips will run cooler. Maybe it wont be up to snuff with Intel.

Now the biggest problem with your filth is that Hammer specs aren't out. system integrators need thermal information to make designs. And when Dell and friends get the thermal data, along with everyone else, then you can talk about what kind of heat output AMD-64 will have.

Everything you say is conjecture, crap, fluff. You don't work for AMD, and judging from your intellect, if you did, they laid you off.

Wrong. You are such a FUD/conjecture machine it's unreal.

The Intel shrinks of the P3 (.13) with the integrated heat spreader are quite cool to the touch under load. In fact, the FIRST CPU Intel shrunk was the Pentium III. You know, you can READ about CPUs. http://www.intel.com/intel/finance/pricelist/ , http://developer.intel.com/design/PentiumIII/specu pdt/ and for P4 lovers, the specification updates for the p4, http://developer.intel.com/design/Pentium4/specupd t/.

Now there is similar documentation available from AMD as well. Now instead of getting anecdotal evidence, (Which can show you clearly there are many holes in this guys Pentium 3 theory), you can see steppings, dissipations, packaging and thermal characteristics for CPUs, not some asshole's rendition of fact into half truths. You want to talk about heat, talk about dissipation first.

Intel could put out a 1.6GHz Pentium III tomorrow [The 1.4 is able to be clocked at that frequency]. They stopped the P3 at 1400 with 512cache for a reason - it started to compete with Pentium 4.

Do you expect anyone to believe you have ever read EETimes or have a sub to Microprocessor Report? I mean, you are poorly summarizing gamer sites here continually.

By the way, for anyone thinking of this guys trash, the Athlon shrinks and the hammer chips will run cooler. Maybe it wont be up to snuff with Intel.

Now the biggest problem with your filth is that Hammer specs aren't out. system integrators need thermal information to make designs. And when Dell and friends get the thermal data, along with everyone else, then you can talk about what kind of heat output AMD-64 will have.

Everything you say is conjecture, crap, fluff. You don't work for AMD, and judging from your intellect, if you did, they laid you off.

Undocumented? Hows about this, this or this. Assembly may be useless to you at the moment, but as the original poster said, learn it and then it won't be.

The document linked below from Intel's website says:

"As a demonstration of compatibility with gcc, Intel C++ 7.0 Compilers for Linux have successfully built and run the Linux kernels, Version 2.4.18, on IA-32 and Itanium processors using a limited number of temporary source patches. The temporary nature of these patches will be addressed in the future either by adding extensions to Intel compilers or by working with the Linux community to replace or reduce usage of less frequently used, non-standard language features in the kernel sources. Both efforts are currently underway."

LinuxCompilersCompatibility702.pdf

And also, you may be interested in my post regarding FreeBSD patches also ;o)

--
Dreamweaver Templates

May I remind you that Intel is not exclusively in the chip market either. Intel spread to new concepts in computing years ago and are better of for it (e.g. From their site: Consulting Services, Compilers, Performance Analyzers, Threading Tools, Training Center, LANDesk* Software etc...) While most of these are certainly related to the PC chip industry it is not nearly as narrow as AMD.

In doing what Intel did years ago, they are actually increasing their competitiveness. In fact a quick look at Intel's (INTC) financials confirm just that.

Hats off to AMD. for keeping capitalism and competitiveness alive.

Non-commercial only. So no good for Redhat, but everyone else is welcome to use it.

Here's another site with videos from both low and high power rockets:
www.videorocketry.com
(videos available Here)

One prime candidate for a low power camera is the discontinued Intel Play Digital Movie Creator. You record video onto a chip and upload to your PC via USB. I believe I saw this camera resurface at Toys R Us under the manufacturer name "Digital Blue". Anyone?

Here is an example of the Intel camera at work in an off-the-shelf Estes rocket with a payload bay added

And here is a rocket with Gumby as the pilot.

It's difficult to estimate, because the human brain is incredibly fast at some things (recognizing a face/voice, processing multiple sounds/images simultaneously, etc...) that would take a computer much longer to do, but on the other hand, it's rather slow at performing specific calculations (How long does it take you to add 100 integers together?).

Even so, the human brain is rated somewhere at millions of gigaflops. Quite interesting. Here are some articles (google for some more):

http://www.intel.com/pressroom/archive/speeches/jt 101100.htm
http://zinos.com/cool/zinos/scan/se=AR002649/sp=vi ew_article/rs=yes/go.html

Even at 733 MHz the Intel 80200 processor dissipates less than 1.3W! This is accomplished using the Intel® Superpipelined RISC Technology-the 7-stage integer, 8-stage memory superpipelined core achieves high speed with ultra-low power consumption.

It looks like you still have to add an external controller chip for devices, but it's probably smaller than the ICH2 and ICH4 found in almost all Pentium 4 systems.

"It's just that Intel/AMD didn't make a song and dance about breaking the GFLOP barrier..."

I don't know 'bout AMD, but Intel has these funny BunnyPeople to promote anything from breaking speed limits to new processors as shown here. So contrary to what you believe, yes Intel does make a song and dance(plus commercial) about [insert_marketing_gibberish_here]!

Intel has a nice tutorial on the subject.
I know that of the Microsoft OS's, only the XP family supports the Hyper-Threading. I couldn't tell you if any other OS's support it.
Distilled down, the processor creates a virtual or logical second processor which assists it in using underutilized resources.
A lot of multimedia vendors would be interested in this, a lot of gaming vendors will jump at this.

And here is the link to Intel's view on this...