Slashdot Mirror
Tom's Hardware End of Year CPU Roundup
Wister285 writes "Tom's Hardware has just posted one of their now famous CPU comparisons. Aside from looking at all of the nice graphs, they also compare the speeds of overclocked processors with their factory rated counterparts. It looks like the AMD chips just don't overclock as well as the Intel ones do, but when run at their specified level AMD almost always has the best price/performance ratio. Hopefully the upcoming year will be as promising in the processor sector as 2003 was!"
there is no PPC 970 on there.
I am the Alpha and the Omega-3
Anandtech is generally the best place to find information on anything you're looking for and is where all the cool kids go. They go above and beyond the call of duty in all of their reviews, and their monitor reviews are unsurpassed.
This linked article is an excellent roundup of the ongoing battle between AMD and Intel. It holds a lot of insight for people who have not been following the news closely.
However, it has to be pointed out that he missed several important incidents:
- AMD alliance with SUN: news article
-AMDs deal with Tippet studios: We built some prototype desktop workstations powered by AMD Athlon(TM) MP processors. We had tried systems powered by a competitor's processors, and they worked fairly well. However, we absolutely preferred the performance of the AMD Athlon(TM) processor. A good part of the advantage comes from the performance of AMD's floating point engine, which is very important to compute-intensive operations such as rendering.
-Intels new challenge in process technology with a cheap strained silicon process, finally unveiled at the iedm. AMD, this will be a touch one: IEDM article
Sure, a Mac is a Mac but there should be a G5 performance comparison with there. After all, not too many Tom's Hardware readers have Itaniums in their home PCs. And with the PowerPC970 (G5) climbing to 3Ghz by March 2004, it should really be included in the article.
If at the very least, they could do speed comparisons on the AMD64, the P4, and the G5 all running various Linux distributions to make it fair. (I'm heavily assuming the Yellow Dog distribution supports the G5)...
"Right now, somewhere in this world, Scott Baio is plowing a woman he doesn't love," - Peter Griffin, *Family Guy*
A few other popular sources of information include:
HardOCP
Dan's Data
X-bit Labs
Ars Technica
Regards,
--
*Art
Overclockability reviews are pointless for a couple of reasons. The first, of course, is that there are never any guarantees - not every one of the famed 300MHz celerons would run at 450MHz, and just because the few samples a reviewer tests overclock well (or poorly) does not mean that all chips will be similar.
The other major problem is that review parts are often hand-picked, nullifying their value as indicators of overclockability completely.
My server
Well, My xp2100+ (1.733Ghz) is currently running at close to 2250MHz (195fsb, 11.5x Multiplier), which is over a 500MHz OC.. I'd call that a pretty nice OC, as it's still only cooled by air. . .
I hate to feed the trolls, but the sibling speaks the truth. This poster, rkz, is not a troll, but he is recycling comments. Not to mention his evil .sig
Yawn.
He has missed one very important fact. Very few of us need any more power then a 2.5 gig CPU. And INTELs 2.5 is twice the cost of AMDs 2500. I run better then 100 FPS in any game that I want to play. Including such hogs of power as BF1942 with the DC mod....
Just because it's the same OS does not imply that it will be a direct comparison -- they are completely different archetectures and can never be directly compared; an indirect comparison is about the best you can do. Still what does it mean? Not a lot, since the archetectures are fundamentally different.
It would be a waste of THG's time when the whole idea was to compare x86 CPUs. Yeah it ignores the PPC -- why? x86 archetecture comparison is an apples to apples comparsion.
Karma whorin' since 1999
Who bothers to overclock a CPU anymore? With the falling prices of machines, you can almost replace it for the same cost. And 2 CPUs are always better than one, because you can run them in parallel.
Don't forget Tech Report
The "thermal problems" with the AMD Athlons is a PERFECT example of why you should NOT read Tom's Hardware Guide! At the very least do not take the articles read at face value without verifying the facts first!
1.) Their P4 was shown to run at a constant 29C. Thermal throttling on the P4 doesn't even start until ~65 or 70C. If the chips were running at 29C, they wouldn't be throttling at all.
2.) The P4 can throttle down to an absolute minimum of 1/8th of it's clock speed, though it's set to 30-50% by default (factory setting) according to Intel's thermal design guidelines. At 30% of it's clock speed, a P4 will still consume easily 20-30W of power, which is WAY more than you can disapate with no heatsink. Yanking the heatsink off a P4 WILL cause it to crash in a very short period of time.
3.) The comment that was made that AMD's thermal sensor could only react to 1C/sec temperature changes was absolutely ridiculous and CLEARLY showed that the author was completely clueless! Such terrible performance couldn't be accomplished by incompetance along, you would really have to TRY and make it that bad!
The whole deal about the instabililties of the PIII 1.13GHz wasn't so much technically incorrect for the simple reason that there was next to no technical info provided, it was almost all just self-congradulation.
I DO judge the articles by themselves, and the articles on Tom's site generally leave a LOT to be desired. The article linked from this story seems to be mostly fluff with a few benchmarks requiring the standard (ie very large) grain of salt.
so what exactly then is the fastest solution? they dont exactly specify that at the end of the review. i've got some x-mas money to spend, and I'm not sure weather I should buy a AMD 64-bit chip (to prepare for the onslaught of 64-bit software) or to buy the latest p4 chip? I'm looking for the fastest solution and a solution that will carry me the longest time (at least a year and a half)
What I would like to see - "If I'm going to overclock, which one pays better"?
First they give overclocking capablities and then non-overclocked price/performance ratio.
We know Intel CPUs are overclockable better but more expensive than AMD.
So, say, I can buy a 2GHZ AMD and overclock it by 300MHZ, getting 2.3GHZ. For the same money I can get a slower Intel and overclock it more. Now, if it was that I can get i.e. 1.7GHZ Inter and overclock it by 600MHZ, it would mean the CPUs are pretty much equivalent for me. Means - about the same price per megahertz overclocked. But if I can buy P4 1.6G overclockable by 500MHZ, giving total 2.1GHZ, it just pays better to buy the AMD.
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
"A few other popular sources of information include:"
;)
Not to mention Ace's if you're really into all the nitty gritty details of things. They do outstanding reviews and technical articles, but can get pretty heavy on the technical details. So far, Ace's is the only place I've found that actually goes over my head from time to time. I do enjoy the challenge.
-- "Government is the great fiction through which everybody endeavors to live at the expense of everybody else."
Has anybody overclocked a Z-80?
In a word: yes!When I was a young and foolish electronic engineering student I and my friends did just that and partially ruined an otherwise perfectly sound rubber keyboard Sinclair Spectrum. I can not remember the exact details but it was not a succesful project. IIRC we tried feeding the system clock line from a squarewave of our own making and tried to run some timer code in an EPROM to flash an LED on an i/o port. My guess is that the Sinclair support chips (and possibly even the NEC Z80 chip our spectrum used) were like AMD processors: just about able to work at their rated frequency, not higher.
I've not looked at a z80 since then but a quick Google search finds that the instruction set has not faded away, here are just two offerings claimed to be Z80 compatable.p rocessors/
http://www.rabbitsemiconductor.com/products/Micro
http://www.ab-semicon.com/datasheets/181e-20.pdf
I've not tried tandooring a haggis yet, you've given me ideas.
Oxford Dictionaries Online
I would tend to look elsewhere for the stability issues you're seeing. While no product is ever 100% perfect by any stretch of the imagination, the AMD chips, in my experience, don't have any more problems than Intel chips since the Athlons. If you could tell me which configurations you've had problems with, then perhaps I could shed some light on where things are going awry.
Generally speaking, I find that using a name-brand power supply, such as Antec, with a Gigabyte or Asus mainboard, and crucial memory solves virtually all stability issues. You can actually put together a pretty nice system for around $450 - $500 using high quality components. The problem with buying a system that's pre-built is that you have no idea who's making the parts. For the cheaper pre-built systems, it's often an ECS (aka PC Chips) board with generic RAM and a generic power supply. It may work well for a while, but you'll invariably run into problems. Personally, when it comes to servers, I want something that I can just build then sit in a customer's office for a few years without any necessary maintenance. I've had success with both AMD and Intel in this area, and I'm now leaning much more towards the AMDs now that the Athlon64s and Opterons are available.
I may actually have a customer who'll put out the money for a really nice dual Opteron system. I'm very much looking forward to building that, as it'll be sitting on a freshly-built gigabit network when it's completed.
-- "Government is the great fiction through which everybody endeavors to live at the expense of everybody else."
This must be the only site that does not mention the Athlon 64 in the conclusion. Therefore I can only draw one conclusion (if you remember that a Athlon 64 3000+ outperforms a similar priced P4) Tom's hardware is done for.
I've been following Tom's hardware for years on end, and I loved their articles on RAID and drive benchmarks. Nowadays the articles are mostly written by mediocre "editors" though, and they bear little resemblence to articles by Tom himself.
To be fair, sometimes they still have great reviews (printers, screens and harddisks mostly), but you will have to look for them between articles that should never have seen the light of day.
Linux users should avoid this Windows site at all cost.
Weird; I've always felt that AMD was the value leader, if you don't need Genuine Intel for some reason. I may be wrong, but could the difficulty of clocking AMD vs Intel be because AMD is already optimized so much nowdays? I don't have any recent experience with it, but that's what I always thought. Could someone explain to a geek who hasn't used them since the K6?
C|N>K
I just upgraded (MSI motherboard died) from an Athlon 1.33 to an Athlon XP 2600+ (1.92.ghz). Can't tell much of a difference. Seems kind of depressing but then I remind myself that w/ negligible difference between last year's and this year's processors, we can all afford to wait for the 5ghz 64bit processors of our dreams.
This guy is way out there
Anyone who knows about this stuff will tell you that Tom's is notoriously biased. It can be shocking. He has been caught out on numerous occasions - photoshopping pictures of cpus, reviewing certain components on crippled test rigs, swapping colours over on his graphs without telling the reader; you name it, he's done it. On his original A64 vs P4EE review he even benchmarked the A64 with three year old 100mhz SDRAM. Unfortunately, hardware newbies (including /. it seems) don't realise this and take what he writes as gospel. I've lost count of the number of innocents i've seen who have bought second-rate hardware on the basis of a THG review. At the moment he is pro-intel, and slightly pro-nvidia for graphics although this is less marked.
AMD's 1.4 GHz Opteron 240, thanks to being "obsolete" is now down to about $215 per chip. (Anyone who thinks Opterons are expensive, is on crack.) Throw a couple of these into a dual-socket-940 motherboard (about $360), and you will have something that can bite the head off of (and shit down the neck stump of) a high-end single P4 system. And costs about the same (not counting the P4EE, which costs more).
The Pentium 4 "Extreme Edition" is the ultimate ripoff for suckers. $1k for a processor? You can get four "obsolete" Opterons for the same price, which make the "extreme" chip look extremely slow. (Hm.. trying to find a quad-940 mb to look up the price, but I'm failing. I know they exist, and there's no way they cost over $600.)
Of course, you can play the same dirty tricks by building multi-P4 systems out of older "obsolete" versions of the P4 which are cheaper, too. But I think the Opteron still wins. The point I'm trying to make is: "day-old" chips are cheap, and if you build SMP systems out of them, they slay!!
"Believe me!" -- Donald Trump
I'm surprised no-one else is bringing this up ....
The review takes pains to point out that AMD-64 binaries are as rare as hens teeth, and for the reviewer's primary audience who are gamers on Windows, and who have to run whatever P4-optimised or Athlon-optimised binaries the games vendors supply, that's pretty much true.
However, for many readers of this august forum, things are a bit more flexible - the only app I run at home that works the CPUs at all hard is digital video processing (transcode / mplayer / mpegenc on Linux), all the binaries for which are of course built from source, thus could potentially be 64-bit if one had AMD-64 hardware and suitable compilers.
Likewise, for the scientific community using Beowulf clusters, who generally run home grown code, this surely has a lot of potential.
Can someone post a summary of the state of the art in terms of AMD-64 binary output from gcc/egcs, and some info on how well it runs with CPU-intensive number crunching like this?
Professionally speaking, all our stuff at work is Java based, and we are looking for price/performance and space/performance ratios - our latest batch of servers (1U pizza boxes with desktop 2 CPU chipsets are the best price/perf compromise) have dual P4's because of the better memory bandwidth of the i7500 dual channel setup compared the dual Athlon chipsets which were stuck at single DDR-266 for the longest time, but if there was a byte compiler which targeted AMD-64 I could see potential for really nice price/performance with the Socket 940 systems, and even just using 32-bit code the higher memory bandwidth would help a lot with Java apps.
Ummmm.... Tom's yet again incorrectly identified a CPU. IA-32 != IA-64 people, however backwards, IA-32 = x86-64... Of course knowing how perceptive people are on /. this has already been posed, right?
Well, compilers generally stress the memory bus a LOT, and if you aren't multi-threading (make -j3) your compiles your CPU has to wait for your DISK to fetch files to compile.
The most important factors in compilation speed (assuming you're sticking with one compiler) are CPU, bus speed/latency, memory size (for caching), and disk latency. Dual CPUs won't do ANYTHING for you unless you multithread your compile jobs, otherwise 'make' only dispatches one job at a time, and each job can only occupy one CPU.
I've found that recompiling the compiler from source with hand-tweaking helps too. In Gentoo the GCC source builder strips a lot of the flags, I undo that and make sure to use 'safe but effective' flags for a faster GCC.
Also remember that -O2 is actually a lot easier to compile and often faster at runtime than -O3 because of modern CPU caching mechanisms. (-O3 unrolls loops, which isn't much advantage on CPUs with larger L2/L3 caches).
"Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
I used to overclock, but I don't anymore. It mattered when a medium speed CPU was barely affordable, and then I could ramp it up to being a fast CPU by OCing. And then when CPUs starting getting cheap it turned into a hobby, and I'd buy a new CPU not because I needed extra speed, but because I just wanted to see what I could pull off. I had MEGAHUGE fans all over the place and finally graduated to water cooling. I was even starting to think about cryo stuff. Then one day a year or two ago I bought an XP2000+ for $65 shipped. I even clocked it up for a few days, but it was so fast at stock speed I just couldn't tell a lick of difference. Stuff happened either instantly, or instantly. The only delays on my system, were non-CPU related. Now today, for practically no money at all, I can have a rediculously fast CPU, or a rediculously fast CPU, depending on whether or not I want to try to clock it. So I don't bother.