Slashdot Mirror
Comparative CPU Benchmarks From 1995 to 2004
Lux writes "The guys over at Tom's Hardware Guide have been busy recently! They've compared over a hundred different architectures dating all the way back to the Pentium 1 in one huge benchmarking effort. Looking to upgrade an older system? Unlike most benchmarks, which compare modern systems to other modern systems, these charts can help you figure out if the cost of upgrading is worth the speedup or if you should hold off for a bit longer."
I currently have a 486 with an (upgraded) 900MB hard drive, cdrom drive, and a whopping 32MB of ram. And windows 3.1 + dos. What are my upgrade options?
It's easy...if the room is getting a little too chilly for my liking, I upgrade to a faster processor. Problem solved.
A /.'ing that I can actually cheer for!
Artificial benchmarks tend to exaggerate minor differences in speed that aren't noticable or relevant in human time.
The best analysis of whether you should upgrade is a subjective one. Sit down at the computer. Does it do what you want or not?
Benchmarks tell me my Radeon 9800 is horribly out of date and imply its too weak to play any modern games. But I know from experience, that's bullshit.
I don't need no instructions to know how to rock!!!!
I went from a 486 to a Sempron 2500+. Unfortunatly the artical doesn't go back far enough so I can't tell if it was worth it.
"I use a Mac because I'm just better than you are."
The articles not even up 3 minutes and its slashdotted already!
3 minutes, as i'm writing there're 2 comments. This ain't right.
They must've dropped them on the road on the way to the benchmarking lab. :(
If they incluce Cryix, VIA C3, Trans processor. Possibly event a G3 G4 or Alpha and Sun system as references to see how far Processors development have gone instead of making it strictly x86. Then again. Tom's hardware is usually just x86 site.
Whoops, pasted the wrong wikipedia link. The correct one is Intel 486
So YOU'RE the one hosting that site. Fastest slashdot effect ever.
-Matt
--- Need web hosting?
Back when I upgraded my 386 16 Mhz, I told myself that I'd upgrade every 10x in performance gain. I upgraded to a Pentium 90 Mhz, then to an Athlon 900 Mhz. It seems that with the recent troubles of AMD/Intel at breaking the 4 Ghz barrier that I won't keep my 'promise' anytime soon, sadly.
How will they keep their market alive if they can't upgrade the performance? Its not like CPU chips are burning easily anyhow... so why get a replacement if the performance gain is not worth it? (Especially for web browsing / text editing only folks who upgrade based on marketing ONLY... yes! 3 Ghz more will make your internet go faster! Heh)
Eureka Science News - automatically updated
turns out they left the pentium 1 in the server, so it's dead now... they'll figure out the cost of upgrading it later.
Runnin' On Empty
You can get this kind of info elsewhere
Actually, they only benchmark one architecture, x86. A real shame, I would love to see a thorough comparison of *multiple* processor architectures over a long period of time.
I choose to remain celibate, like my father and his father before him.
They haven't been busy recently. They just updated the guide they did quite some time ago. Not very much new to see here...
It seems the server that hosts the /cpu/(stuff) died.. the main site is accessable, as are other parts of it.. Maybe /. should have posted motherboard reviews instead!
The last time I looked there where more distributions available for 68030 than for 486. Just take a look, most distribution except Linux-from-Scratch and variations like Linux-nearly-from-Scratch are build to run on a minimum of Pentium.
"Life is short and in most cases it ends with death." Sir Sinclair
I wonder what hardware they're using to run Tom's site?
Tom uses several of these. He has multiple dedicated boxes with this company (the same one I use, but I am on shared hosting). This host uses FreeBSD, so maybe his site died along with BSD? :-)
24 beers in a case, 24 hours in a day. Coincidence? I think not!
"Comparative CPU Benchmarks From 1995 to 2004"
I only see x86 CPUs. What about the PowerPCs, SPARCs, MIPS, Alphas, ARMs, and so on?
For instance, the m68060 was the first consumer level processor with branch prediction and branch folding, superscalar dispatch, and real-world throughput of more than one instruction per clock cycle. Except for floating point where it performed only modestly, the m68060 seriously outperformed the Pentium in spite of only having a 32 bit data bus as compared with the Pentium's 64 bit bus. Isn't this significant in illustrating the influences in processor architecture?
http://www.sixgirls.org/ is an m68060 Amiga running NetBSD 2.0. Still very useful after all this time. Where are all those Pentium 60 machines?
Part 1: http://www.tomshardware.com.nyud.net:8090/cpu/2004 1220/index.html4 1221/index.html
Part 2: http://www.tomshardware.com.nyud.net:8090/cpu/200
http//injoke.org -- Culling The Interesting
ever.
So says the title of a chart they post.
Note that the CPUs in the chart start at 600MHz. Who knew the old AMD 286's ran so fast!
September 2011: Looking for Cocoa/iOS work in Boston area Cocoa Programmer Quincy, MA
I can't get to the article, so I have to wonder what kind of benchmarks they used that were consistant on all those platforms. Old benchmarks tend to freak out on newer hardware, and I can't imagine newer benchmarks running properly on older hardware.
http://crummysocks.com
I think toms site is going under with all the traffic from /.
.. :) lemme read ..
takes a couple of minutes for a page to load
I was first there peeps
Would be nice if they included what software was available at each level. For example, I had XENIX (from Microsoft and SCO) on an early 386 and it rocked (full 32-bits), but Microsoft didn't do 32-bits on the desktop until Win95 (10 years later), and didn't migrate to a full 32-bit O/S until Win2K and XP (nearly 15 years after the 386 came out).
Why did they use an 8mb video card for the older motherboards that don't support AGP?
Matrox Mystique G170
Memory: 8 MB SD-G-RAM
They should use the fastest availible video card if they are testing CPU speed. My 200mhz pentium pro with a 16mb TNT card ran Quake 3.
that doesn't mean that subjective judgement is "the best analysis", it just means that current benchmarks suck. Subjective judgement is what built the cathedrals of yore. That's because they didn't have a real "best analysis" back then. (And sure a lot of them held up, but most of the crumbled within a few years and in any case they are vastly overengineered, however aesthetic.)
The article is a bit slashdotted but it looks like it doesn't go back all that far.
Just a teaser, I have been running a collection of benchmarks since the Pentium 90.
At the time, I was involved in a huge UNIX engineering workstation benchmark. I felt we needed something more constant than the applications to compare performance (the engineering apps constantly change). So I quickly assembled everything I could find that could be easily run. These are mostly 'toy' benchmarks, but the results are still interesting.
For these int benchmarks, higher is better:
c4.s c4.64 dhry21 hanoi heapsort nsieve nsieve TOTAL
Kpos/sec Kpos/sec MIPS mvs/sec high High Low
MIPS MIPS MIPS
P 90 92.7 94.2 68.6 51.2 43.55 111.0 33.3 494.6
md64b 4050.1 4167.8 4914.3 2708.8 3333.7 3333.7 610.4 21782
Float: Higher is better, except for the fft's.
flops20 fft tfftdp
MFLOPS MFLOPS MFLOPS MFLOPS TOTAL time time
(1) (2) (3) (4)
P 90 13.3 12.8 18.1 23.8 68.0 3.07 16.81
amd64 1120.9 1004.3 1480.9 1834.7 5440.8 0.04 0.42
The P90 was running RedHat. The AMD64 is my new desktop, a 90nm 3000 OC'd to 2430 Mhz. My data also includes systems from DEC, HP, IBM, Sun and SGI. I also ran 10 matrix multiply benchmarks as part of the effort.
I have never gotten around to publishing the results or the collection of benchmarks.. Maybe it is time.
I bought my computer in 1982... how will I know if it's worth upgrading if the data only goes back to 1995?
You see? You see? Your stupid minds! Stupid! Stupid!
Actually I shouldn't give Tom's Hardware a hard time (like everyone else seems to). As articles go, the reviews of high-end ink-jets, the 8-channel RAID6 card and the Viewsonic media center were quite interesting (and a lot more recent than the CPU round-up too).
These days though, my favourite reviewer is Dan (who posts here now and then). Dan seems to understand that a million graphs showing you the statistically insignificant difference between the latest mobos / graphic cards / processors / ram sinks don't really make a great site.
Alot of the benchmarks have the bottom 4 or 5 at zero, ie quake III and RTCW. What specifically got my attention was the K6-2+ 500. This is almost exactly what I'm running (don't laugh) and I've played both of these games on it quite well. They may not be at max graphics settings, but they look pretty good to me. and yes I am planning on upgrading, but I'm waiting on the nforce 4 mb'sso I won't have to upgrade the next time I want a new graphics card.
Life is pain. Anyone who says differently is selling something.
but I'm sure that I read a similar article at tom's this year comparing x86 CPUs from ~386 to recent systems.
IIRC they decompressed zip files and encoded video stuff among other things. It's quite impressive to see a fairly modern CPU performing the same task in minutes which used to take several hours on one of those 'antiques'.
I don't read replies by ACs.
I think Tom's Hardware servers need an upgrade...
I think its a measure of the true utility of the article. People are actually reading the article BEFORE posting, a rarity on Slashdot...
You are in a maze of twisted little posts, all alike.
The 486 has had a built-in math coprocessor ever since it debuted. After the 486 was around for a while, they made a stripped down version without a coprocessor called the 486SX. The plain 486's were called 486DX.
You could get a coprocessor for the 486SX, but not the DX. From what I've heard, the original 486SX's were actually re-badged 486DX's whose math coprocessor unit was either not functional or just disabled. When you bought the 487SX "co-processor" you were actually buying a fully functional 486DX that disabled the other CPU on the board.
whatever it is, it sure isn't /. proof.
Does anyone else have the sudden urge to run off to Kinko's and get the giant master benchmark charts printed off as posters?
...Or is that just me? : - )
If you installed Intel's PII Overdrive for Socket 8 mommyboard you could obtain MMX, but that was the only way.
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
Looks like they should look at the chart themselves and decide whether to upgrade the 8086 they have running the webserver...
In the meantime the model in the holiday buying guide is hot.1 5/inde x.html
http://www.tomshardware.com/consumer/200411
However, beware of the 486DX50 vs the 486DX250. The 486DX 50 was a true 50Mhz part whereas the DX2 were only 50Mhz internal to the chip with the bus running at 25Mhz. Same thing for 486DX2 66's. Most programs ran slower on them than a trus 486DX50 due to the slower (33Mhz) bus speed.
Yup. Then Intel had to confuse the issue by releasing the 486DX4. Just as the DX2-50 had a 2x multiplier with a 25 mhz bus and a 50 mhz core speed, you'd think the DX4-100 would have a 4x multiplier with a 25 mhz bus and 100 mhz core speed. But it was actually a 3x multiplier, with a 33 mhz bus speed. They should have caused it a DX3
Though that was before my time, I do know for sure that AMD made processors way before they got the clock speed to 600MHz, so you're right about that.
When you see better computers than the one you're using at Goodwill/Salvation Army/St. Vincents .. it's time to upgrade.
I've got a Duron 1200 and am trying desperately to upgrade to an nForce2 / XP 3200+.
After a bad board and bad processor (from partspc!) blew a RAM chip I had bought for it, I ended up RMAing all three of them, likely with little or no compensation coming my way.
So here I am, after Christmas, stuck with one functional 512 stick of PC3200 Corsair after have spent $400 or more on my upgrade fest.
That's why I don't upgrade very often, besides the lack of money. Because I know that there's a good chance whatever I get isn't going to work as well as the old parts, thus stability is a good thing for me.
Job? I don't have time to get a job! Who will sit around and bitch about being broke and unemployed then?
First Link
Second Link
People don't really detect a perceivable performance boost for less than 25% more actual performace. And the perception of that 25% varies wildly from 'fneh' to 'wowieeeee'.
But I'm talking to a bunch of gamers for whom that last 0.4% performance boost is worth more than a hot cheerleader full of X.
Sure. I had the exact opposite problem... I went looking for the graphs, and couldn't find them because I wound up in part 1. :)
h tm l
http://www.tomshardware.com/cpu/20041220/index.
It's important to note the subjectivity factor in benchmark measurements. Too much focus is put the raw numbers, though that's easy for the marketing folks. There's a good column by economist Alfonzo Wallings that covers this kind of thing (Forbes I think, don't have the URL on me at the moment).
Dude, the article is mostly pictures, very little reading going on.
Bad boys rape our young girls but Violet gives willingly.
The primary end-result of the evolution of this commodity hardware is the fact that expensive software is now just obsolete - plain and simple. Ten years ago, there was a justified price premium associated with state of the art software algorithms. I still see these zealots for the DB companies raising these red-herring issues as to why every organization should still spend hundreds of thousands of dollars on expensive DB software. With the evolution of hardware the way it is, any credible SQL DB Engine could run almost any company. Same thing for web and application servers. How can anyone justify paying for these things when the hosting companies prove everyday for thousands of tech-savvy companies that the free solution is just as scalable and more secure. I used to maintain several Windwos servers and finally ported them to a hosted Apache solution - for about 5% of the cost. The sites are always available and the admin tools are web-based and better. And I don't have to hire these guys that want to go to the MS Marketing summit for a week every year so they can continue to administer the "low-cost MS solution approach". If you have any hesitation, make the switch. This stuff is now public domain, don't pay for this stuff. Those days are long gone. In the new model, it only makes sense to pay for software that solves industry-specific problems - not for tools that cost a fortune to maintain and invite tech companies into your business to meddle and start religous wars among the employees. www.SoftwareObjectz.com
http://www.softwareobjectz.com
But I'm talking to a bunch of gamers for whom that last 0.4% performance boost is worth more than a hot cheerleader full of X.
These are the same people that get into squabbles over which display adapter is better when A gets 150FPS (frames per second) and B gets 140FPS but with better double-dodecahedron rendering or whatever. It makes no difference to them that most human eyes cannot distinguish between 40FPS and 80FPS, let alone anything above 100. However, if the [abstract and wholly meaningless number] is higher then the product must be better. Oh, well - at least the cheerleaders are safe.
I want to drag this out as long as possible. Bring me my protractor.
You can't even run it on the lowest CPU's they tested.
Why not use GLQuake or even *gasp* software Quake?
My wife responded to my description of the SX, DX situation with "So, you want the 486DX, not the 486 Sucks, right?"
1333 4.02 PowerPC 7457 (1.5 year old 17" powerbook G4)
HIV Crosses Species Barrier... into Muppets
If you take a look at the big chart, why is it that the clock speeds have increased by ~10 fold from start to end, but the front side busses have only increased ~5 fold?
What i find a really snazzy card would allow me to do is run with ANISO and FSAA up really high, maxed out textures, etc. No guarantees but i think upgrading your mobo/cpu/memory would have a greater effect. Just as you said, when you increase players it drops like hell. This is obviously not video card related.
----
Go canucks, habs, and sens!
Has to say about benchmarks. Especially considering that they only accept advertising money from AMD.
End of Line.
Unless there's a late-release unit that I'm unaware of, in which case I will sit corrected, no Socket 8 Pentium Pro supported the MMX instruction set at all. Their release predated MMX.
I own seven PPro boxes myself, most of them with stepping level 7 or 9 CPUs, and one with a PII/333 OverDrive chip installed.
It's fairly trivial to demonstrate that MMX instructions are not present -- almost any CPU identification software will test for that, and the one Linux game I've been interested in which has a hard MMX requirement will fail to run on all of my PPro boxes except the one with the OverDrive.
I suspect you either misunderstand the issue being talked about or are actually using a chip which is not a Pentium Pro CPU. An older PII, perhaps?
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
However, beware of the 486DX50 vs the 486DX250. The 486DX 50 was a true 50Mhz part whereas the DX2 were only 50Mhz internal to the chip with the bus running at 25Mhz.
Of course, then there's the other side of this issue, which was VLB: Vesa Local Bus (which is what we used for fast access to graphics cards [and sometimes IDE disk] before PCI/AGP) was only spec'd to run at 33MHz tops, and many VLB cards wouldn't run at 50 MHz. So you often had a choice: buy the 50 MHZ 486 DX to get the full 50MHz bus speed, but use an ISA graphics card, or buy the DX2/50 and use a VLB graphics card a 25MHz. The usual answer (once it came out) was the 486 DX2/66, which was a 66 MHz processor on a 33 MHz bus, topping out the local bus clock. Some things still ran faster on the 486DX/50, but games usually ran better on the DX2/66 due to the higher graphics card throughput.
Of course, the highest end graphics cards (Diamond's Viper, ferinstance) could generally handle a 50MHz clock. But most of us didn't have that kind of scratch lying around at the time.
-JDF
which are fundamentally flawed for real user needs. They compare different systems running the same software (can you believe?) That might be relevant in academia, but in the real world, it's much more useful to compare complete systems: processor, disk, OS and applications. When you do that, you'll find that for the bulk of common user applications (web browsing being a notable exception), today's systems simply aren't much faster than older ones. Compare a system running Windoze XP / Office XP running on a 3 GHz machine today with a Windoze 98 / Office 98 machine running on a Pentium II. Not much difference in usable speed (which I define as time to boot, time to launch an application, time to do a common task like write a letter or create a spreadsheet, etc.) Sure, recalc on a zillion cell spreadsheet will be faster, but two things work against improvement: 1) both machines spend most of their time waiting for the user to do something, and 2) bloatware.
"National Security is the chief cause of national insecurity." - Celine's First Law
Its so nice to see that the sse2 results were used instead of the raw x87 scores for the intel p4 cored chips http://www.tomshardware.com/cpu/20041221/images/ch art_007.png
Viral software licensing is not freedom, it is in fact GNU/Socialism.
2x 2000 2.37 PowerPC G5 I think the parent's observations about this being misleading and varying system to system, OS to OS are very true.
...
benchmark
This is just looking at core cpu performance, not bus and external cache issues.
I am always doing that which I can not do, in order that I may learn how to do it. - Pablo Picasso
Doah! Sorry about the skanky link in my last response. Here's a correct one.
Been an AMD man (not counting laptops), and have had the same main computer now since 1991...much in the same way a lumberjack can claim he's had the same axe for 13 years but replaced the handle 9 times and the head 14 times.
~1989: Intel? 8086, 640k ram, ?video
~1991: AMD 386 DX 40, 4 megs ram, 1meg video
~1994: AMD 386 DX-2 66, 5 megs ram, 1 meg video
~1996: AMD 486 DX-4 133, 5 megs ram, 2 meg video
~1997: AMD K-5 166, 16 megs ram, 2 meg video
~1999: AMD K-6 300, 16 megs ram, 4 meg video
~2001: AMD Duron 600, 128 megs ram, 16 meg video
~2003: AMD Athelon 1900+, 512 megs ram, 64 meg video
Still have the 1900+, but added a new video card with a quarter gig of ram on it. It simply amazes me how much things have changed. I have a laptop now that theoretically should outperform my desktop (except with games). Though looking at that time schedule, I think it's probably time to upgrade my main PC.
The Internet is generally stupid
... as claimed here (presumably as viewed from the desktop)?
I can just imagine PHBs cancelling orders for Opteron servers because "Tom says they don't make them any more".
"There is nothing so simple that works so well that it can't be made to work better by making it more complicated" - ?
Here's a picture of an AMD 286. Real close up.
t ml
http://micro.magnet.fsu.edu/chipshots/amd/286.h
"The Advanced Micro Devices (AMD) 286 microprocessor, fabricated and distributed under license from Intel, utilizes identical architecture and the source code developed for the i80286 chip. AMD boosted Intel's original 6, 10, and 12 megahertz clock speed versions, which are powered at 5 volts, to standard models featuring higher speeds ranging up to 20 megahertz. During the mid-1980s, AMD launched the microprocessor clone market with their 286A processor, which followed their in-house version of the 287 math coprocessor. Released in 12 and 16 megahertz clock speeds, the 286A model did not break new ground, but did feature expanded memory specifications (EMS) and the ability to reassume protected mode, neither which was supported by the Intel 286."
September 2011: Looking for Cocoa/iOS work in Boston area Cocoa Programmer Quincy, MA
yep, I was lucky back then, my vesa local bus graphics card (sorry forgot which one) did work at 50 MHz. So I had the fastest computer of the whole lab, runing 50MHz all over, Yes! It was running OS/2 with a Gopher server and later my first web server. With all pages handcoded in HTML 1.0.
Are the temperature specs posted in this picture really accurate ? That's interesting, because I'm about to upgrade an old pc. The currect case is ok, but it has mediocre ventilation, so I was going with a slower processer (say an Athlon 1800+) to reduce heat. By these specs, I'd be better off with the Athlon 2800+ (which was the one I wanted anyways...)
crap...
my boss is watching me.
Pentium 233 MMX can convert MPEG2>DivX in 113 minutes while converting to XviD only take 59 minutes. cut it down nearly by half.
On the other hand P4 570 (3.8GHz) take 1:40 minutes for a DivX and 2:23 minutes for XviD. Almost double.
What is that mean? Is the CPU architechture very improving?
-=-=-=-=-=-=-=-=-=-=-=-=-=- If picture worth a thousand words, how many megapixels is it? -=-=-=-=-=-=-=-=-=-=-=-=-=-
1/10 th as fast. Please comprehend what you're reading before you pick nits.
± 29 dB
I don't want to start a flame war, but I belive you are referring to threads. One processor can handle multiple threads just fine, and two or more usually better. By adding one or more processors to Windows, it will by no way "fix" the lock up issue. It might make the system perform better under certain loads, but if you have a bad video driver (ATI), it will still take down the entire system. If you have a bad network task (something trying to map a drive to a bogus address) then your system will still crawl down. Very Very Very Very (~Very) seldom does one thread/process take up 100% CPU and STOP you from killing it in Windows or Linux. However even in that case, it would still be possible for someone to write a bad app that spawned a bunch of threads that did the same to all your processors.
The more I learn about science, the more my faith in God increases.
Yet you were wanting him to show proof that a 386 is 10x as fast as P90, showing a startling lack of comprehension.
Any proof he would likely give would result in your further confusion.
± 29 dB
Any K6 can use SDRAM, ie anything a Pentium 3 can use, so can a K6-2. It may be limited to a 66 or 100 mhz fsb, but that doesn't mean a standard PC133 stick won't work perfectly in it. I still use a K6-2/300 as a very decent file server - 3 simulataneous users pulling / putting files on it + serving media files from a raid array (the regular linux software kernel one!) works fine.
1919 2.18 AMD Athlon-XP 2600+
266 24.15 AMD K6-II 266
25 200.22 TI MicroSparc (Sparcstation LX)
Most of the benchmarks are gaming and thus more dependant on the graphics card. Given that not all computers are using the same graphics card, that makes them somewhat unfair.
More interesting are thus the non-synthetic non-graphics application benchmarks, but unfortunately theres just four of these.
What might be interesting with the graphics benchmarks would be to outfit each computer with the state-of-the-art graphics card of its release date. Also unfair, but given that the graphics card used in the test is not uniform anyway (having a somewhat sudden cut-off for computers with only PCI and not PCIe or AGP) it might be a slightly more interesting setup.
Good point, but why can't the improvement be done at the hardware level that's transparent to the users and developers.
-=-=-=-=-=-=-=-=-=-=-=-=-=- If picture worth a thousand words, how many megapixels is it? -=-=-=-=-=-=-=-=-=-=-=-=-=-
The differences are not that dramatic if you are reasonably efficient. I use a (heavily) upgraded K5/166 with 48MB RAM, 40GB drive (BIOS won't see it correctly, but hda=x,y,z solved this issue with linux), PCI TNT2 with 32MB, SB Live! and a Plextor Ultra-SCSI CDROM on an adaptec controller. I also added a 3com 10/100 Ethernet and this thing rox0rz.
I use it to run linux and it is quite adequate for latex text editing. The system is snappy enough for emacs and xfce. Only when you start an MP3 player do you realise that media decoding can eat up ~30% CPU (there are huge differences between various players, of course). I also record some 44.1/16b music from analog and use it as a multimedia box. It can play VCDs too.
As for network use, well, rsync (encrypted) and ssh are really heavy but plain ftp runs pretty well either as a client or server. This machine will easily saturate a 10MBps line with unencrypted static content so it is not that bad for home use as a backup server/multimedia box. Network throughput peaks at about 70MBps if the machine is idle.
P.
P.S. For comparison, my other system is Athlon64 3200+ with 1GB CL2 RAM, 9800Pro and 160GB SATA.
While this is certainly interesting to see, what I like the most about it is that it actually proves the insanity of the Megahertz race. Using the P3/600Mhz as my low end and P4EE/3460 as my high end, I am not noticing any significant instance where the expensive whore is faster by at least an order of magnitude.
Let us not forget that it's not just the CPU but the whole supporting cast that has vastly improved over the years. Even with all the improvements in non-CPU technologies such as RAM, chipset and FSB, a P4 Gallatin CPU running at 3460 MHz does not perform 6x (3460/600=6.06) faster than a P3 Coppermine CPU running at 600Mhz in most cases (not counting the SIMD benchmarks that rely primarily on SSE2, etc.).
So what does this mean? It pretty much means that the great majority of us are buying marketing hype. At retail.
Must-not-watch TV!