Intel 6xx Series Reviewed and Benchmarked

sebFlyte writes "It's been a long time coming, but Intel's first 64-bit desktop chip (the 6xx series) is here now, and thanks to ZDNet it has been thoroughly tested. The article has the full specs of the new family, explains the benefits of the changes, and also the results of tests on the new chips to establish perfomance boosts for games, photo manipulation and video work, among other things."

Comments

Naming

[fembots]

Looks like Intel's really moving away from marketing its GHz. Its Pentium M has been using 7xx, and this 64bit is on 6xx.

Re:Naming

[GeorgeMcBay]

Moore's law will have to be fulfilled in new, unique ways.


Moore's Law was never about speed in gigahertz, but rather about transistor counts, so it'll continue on as it always has, since more cores means more transistors.

Intel plays catchup

[dmf415]

Intel has fallen behind as the mainstream CPU manufacterer. I wonder how long consumer will pay a premium for slower Intel CPUs

Analysts' assessments came after Intel announced last week that it was scrapping plans to develop a 4-gigahertz version of its Pentium 4 chip. On Tuesday, AMD unveiled a pair of advanced new chips, the Athlon 64 FX-55 and the Athlon 64 4000-plus, which analysts believe will be more appealing at the high end of the PC market.

more here:
http://www.newsfactor.com/story.xhtml?story _title= Analysts--AMD-Beating-Intel-in-Chip-Design&story_i d=27767&category=hardware

mostly 32-bit benchmarks...

[Master Bait]

I'm not impressed. All those 32-bit benchmarks to benchmark their 64-bit CPU. Last week Linux Hardware benched the new Pentium against Opteron with real 64-bit apps on a real 64-bit OS.

I love this statement

[dot_borg]

"...although there is, as yet, no shipping 64-bit Windows operating system, which is necessary to make the most of a 64-bit CPU."

I've been using 64 bit Gentoo for like a year now, but I guess that doesn't count.

Slower? Says who? You?

[WIAKywbfatw]

While I agree with you that Intel is playing catch-up in the desktop 64-bit arena, if you had RTFA then you would have found that it concludes with a performance summary that suggests that Intel's 64-bit CPUs more than hold their own:

As far as performance is concerned, the 3.6GHz Pentium 4 660 can hold its own against its main competitor, the Athlon 64 4000+. The Intel chip performs particularly well if several tasks are running at the same time; under these circumstances, the Pentium 4 can outpace its AMD rival even if the latter is quicker at performing the tasks on their own. Thanks to HyperThreading (HT), the Pentium 4 distributes processing tasks across two virtual cores, resulting in more efficient utilisation of CPU resources. Such scenarios are found ever more frequently in the real world. For example, no-one should venture onto the Internet without firewall, antivirus and anti-spyware protection. These services are constantly active and need appropriate resources. Likewise, operations such as data encryption or hard disk defragmentation can load the processor, while the user compresses streaming video or audio data. Under such usage patterns, the advantage of HT is particularly apparent.

With the 6xx-series Pentium 4, Intel has ensured that important functionality such as protection from memory overflow, power management and 64-bit support on the desktop is no longer an AMD domain. And with HT support and SSE3 instructions, the new Intel processors offer additional benefits. One change will annoy Intel, however: on the installation CD for Windows XP Professional x64 Edition, the most important operating system files are no longer in the 'i386' folder; Intel systems must load the installation files now from the 'AMD64' folder. Although that might hurt the industry leader, Intel may draw comfort from the fact that it has already sold more 64-bit chips than the inventor of the x86-64 architecture, AMD.

Now, I have no doubt that Intel's 64-bit offerings will fall behind their AMD equivalents when it comes to bang-per-buck but that conclusion seems to suggest that Intel's chips will still have plenty of bang, as much as if not more than AMD's chips.

Remember, Intel's chips are just getting to market, whereas AMD's have been out there for at least 12 months. Who's to know who'll be lording it over who in a year or two when the 64-bit market is finally more than a tiny subset of the market as a whole? It might be AMD, it might be Intel, or it might be neither.

And before the accusations start, no, I'm not an Intel fanboy or shill who's interested only in running down AMD: this post is being typed on a AMD Athlon-based PC, my last PC had an AMD CPU, and my next one almost definitely will too. What I am interested in is a fair and accurate comparisons of the processors of the future, regardless of who they are made by.

Re:Slower? Says who? You?

[i41Overlord]

That is just damage control going on at Ziff-Davis. Intel used to be a big investor in ZD, I don't know if they still are.

I remember when I worked at a mobo manufacturer back in the late 90's, ZD WinBench was the major benchmark used to measure the performance of chips/motherboards/PC's. This was during the time that AMD came out with the K6 processor, and had a lead in most benchmarks. Next thing you know, Intel's investing in ZD and ZD comes out with an "updated" version of the program. I test it out and suddenly AMD's chips seem to be choking on the benchmarks, while Intel's chips got a nice speed boost. Very shady tactics.

I dealt with Intel and I'll tell you that they're a very dirty company to deal with. They use the same tactics as the Microsofts, Walmarts, and other big companies that use their current power to stifle competition.

Re:Slower? Says who? You?

[Glock27]

Such scenarios are found ever more frequently in the real world. For example, no-one should venture onto the Internet without firewall, antivirus and anti-spyware protection. These services are constantly active and need appropriate resources. Likewise, operations such as data encryption or hard disk defragmentation can load the processor, while the user compresses streaming video or audio data. Under such usage patterns, the advantage of HT is particularly apparent.

This is basically BS. Firewally, anti-virus, and anti-spyware are either 1) dedicated hardware or 2) run sporadically without being CPU-bound. How many people defragment their disk other than at 3:00 AM Sunday morning? And "data encryption" is a non-starter for most except for SSL encryption.

Nice try, though...

That said, dual core and/or regular timeslicing will work for all this as well anyhow...

Re:Slower? Says who? You?

[Jollyeugene]

This article also compares Intel's announced CPU's with AMD's CPU's that you can actually buy. This is just a fluff piece-- same as the "Extreme Edition" that would overheat and that no one could buy for around 6 months.

If you look at yesterdays news, you will see that AMD is releasing dual core chips soon. So when both of these new chips actually are available in quantity-- then lets do a review. Hyperthreading will not look so good then.

Hyper Threading is an engineering solution to try and fix the problem created by Intel's marketing department-- when the company let them design the Pentium 4 to scale on Megahertz and not on performance. After that fiasco, Intel got its butt handed to it on just about all benchmarks by the Athlon. Intel management then panicked, and Intel's engineers salvaged the long processor pipeline with "Hyper Threading".

Dual cores (SMP) are the better solution. When dual cores come out, hyper threading looses most of its advantage.

That article reads like a typical ZDNet article.

[i41Overlord]

ZDNet tries to be slick about it, but there's no mistaking who has influence at Ziff-Davis.

When AMD comes out with something, it usually doesn't make the news or is seriously downplayed. However when Intel fires back a few months later, it's suddenly big news, worthy of headlines.

When Intel releases a CPU that is faster than any of AMD's offerings, there is usually a big story that accompanies the event, praising Intel and their engineering prowess. However, when AMD releases a CPU that outpaces the Intel offerings, the fact is heavily downplayed. You can tell there's some damage control going on at this "objective" media outlet.

Re:AMD64 Inside

[Rosco P. Coltrane]

Ha, like Microsoft will leave that called AMD64. Expect some diplomacy and a renaming. Not that anyone but techies care.

Why would they care? When Intel invented the i*86 line, everybody software manufacturer called any Intel-compatible CPU "i*86" somewhere or other, and neither AMD, Cyrix,... complained about it. Now the situation is reversed: AMD took the lead on that particular 64 bit design, and Intel is just a follower here. It sounds rather normal and deserved that any AMD64-compatible chip should generically called "AMD64" after all.

Re:heating

[Rosco P. Coltrane]

100% of power consumed by a processor is converted to heat. 100%.

So I guess the amount of electricity that flows out of the CPU to the different busses comes from the hamster pedalling inside the CPU?

Re:that is a typo

[Master Bait]

Did you read the article? All cpus are tested with both 32bit and 64bit code.

Only two 64-bit apps, Povray and Panorama Factory. As a matter of fact, didn't you also notice that the benhmarks were mostly SSE math on very large datasets?

This one is a laugher:

The Intel chip performs particularly well if several tasks are running at the same time; under these circumstances, the Pentium 4 can outpace its AMD rival even if the latter is quicker at performing the tasks on their own.

... meaning they were doing SSE math on two very large dataset files. They likely ran Intel-picked benchmarks which show that Intel's SSE instruction set is faster than AMDs. The benchmarks on streaming data hide the horrific latency of DDR2, don't show anything about integer performance, performance of the system in handling interrupts, etc.

Re:What? No more GHZ?

[jacksonj04]

What people also fail to grasp is that MHz make very little difference at all in most applications. My current machine has a 1.8GHz CPU (AMD if you must know) and beats the latest 3.6GHz into the ground. Why? Because the mobo is high quality, I've got oodles of fast RAM (God bless Crucial), my HDDs are fast...

Your processor can be as fast as you like but without the supporting hardware it does nothing special.

Intel is expensive?

[slackmaster2000]

Last time I looked (just now!), Intel processors are only $20-40 more expensive than comparable AMD processors. The big price difference isn't so big anymore.

I'm no Intel fan...I find it silly to be a "fan" of a corporation. In fact two of my three current systems are AMD based, including the one I'm writing this on. However, I must say that buying a machine with an Intel chipset is considerably more idiot proof than buying AMD - mostly because there are just too many AMD chipsets to choose from, many of which (esp. from VIA, SiS) turn out to be pretty quirky. I'm glad that I've found at least some consistancy from nVidia (nforce2 and >).

It's fun to hate Intel because they're the giant, but I for one still have not forgiven AMD for the K6-x processors and all their marketing BS that amounted to outright lies in my opinion. I also think they're stretching it a bit right now with their current "intel comparable" numbering scheme.

Good, fast, reliable systems can be made with both AMD and Intel at this point, and the total cost difference is really minimal. I certainly find strange that anyone would be bashing either company in this regard at this time.

Intel 6xx heat is still a problem...

[dtjohnson]

The Intel 660 reviewed in the article has closed the gap quite a bit with the Athlon 64 4000+ but it looks like heat is still a problem for Intel, even with the new 6xx 'prescott' cpus. The Intel 660 has a thermal design envelope of 115 watts vs only 89 watts for the Athlon 64 4000+. More significantly, the Intel 660 has hardware and software mechanisms to automatically reduce the core voltage and clock speed when high temperatures occur as described in this review:

• Thermal Monitoring 2
  TM2 is overheat protection, controlled by the processor's PROCHOT signal (processor hot). This signal is activated if the CPU's thermal diode detects critical temperature levels. TM2 will dynamically reduce core voltage and clock speed in order to cool down the CPU.
• Enhanced SpeedStep
  SpeedStep does exactly the same as TM2, with the difference that it is initiated by the operating system. Whenever the system load is low, Windows XP SP2 will cause the CPU to lower the clock speed in 200 MHz increments by using ACPI mechanisms. Again, this is performed dynamically, which means that executing a demanding application will cause the system to speed up again.

These sorts of protective mechanisms mean that your "3.6 Ghz" Intel processor might not be operating at either 3.6 Ghz or the core voltage that you have selected when either the motherboard or the Windows XP operating system determine that the thermal situation is getting out of hand. The original idea behind the early implementations of this stuff was to protect your cpu from catastrophic damage when the cooling fan failed, or some similar catastrophic event. Now, however, the stealthy way that Intel is implementing these mechanisms in both the hardware and the OS suggests that they have moved into the realm of selling performance that the system may not thermally be capable of delivering on a sustained basis. Will the user see any indication that their system is slowing down or that the voltage has been decreased? In effect, the Intel systems are becoming more like 'dragsters' that are capable of short periods of high acceleration and speed but perhaps unable to operate at speed on a sustained basis.

Power consumption

[D. Book]

IMHO, the area in which AMD has really distinguished itself is in the power consumption of its desktop processors. Generally speaking, non-Intel x86 CPUs (from AMD and Cyrix) historically had a reputation for running extremely hot. The situation has been reversed in recent years, with Intel pushing the upper limits of power consumption with its Pentium 4 (especially with Prescott) while AMD was doing the exact opposite with its Athlon 64. This has resulted in a huge difference in power consumption between the two competitors. Consider the following CPUs which are basically direct competitors (roughly the same price):

Power consumption at idle
Athlon 64 3500+ (Winchester): 13.4W
Pentium IV 640 (Prescott 2M): 35.4W

Power consumption at full utilisation
Athlon 64 3500+ (Winchester): 47.5W
Pentium IV 640 (Prescott 2M): 129.4W

Source: 90nm Processors from AMD and Intel Pentium 4 6XX.

The often trivial differences in performance look rather insignificant in comparison. Also consider that these results come after Intel's best attempts at reducing the P4's power consumption (enhanced idle states in P4 5XX and SpeedStep in the 6XX) and you can see how inefficient the architecture is in this regard.

This of course applies to desktop CPUs. Intel redeems itself somewhat with its Pentium M in the notebook market.

Techreport's test is from feb. 20th

[Daath]

Techreport's test is from feburary 20th, and shows somewhat the same - AMD64 is king of gaming, and doing one thing at a time - Intel is good at encoding etc, and thanks to HT, doing more than one thing at a time - I would prefer AMD for now (I got a Prescott 550 3.4GHz - I wish I had an AMD).
Ain't it great with competition?! :D