Current Processors Tested With Linux 330
Happy-Jollies writes "The team at LinuxHardware.org have kept us up-to-date on the latest in processors for quite some time now and they're at it again. With the latest release of the Pentium 4 'Prescott,' many Linux users will be deciding where to spend their money. LinuxHardware.org's round-up takes a look at the Prescott, Northwood, Extreme Edition, and the AMD Athlon 64."
Why did they leave out ... (Score:5, Interesting)
Re: (Score:3, Interesting)
Re:Why did they leave out ... (Score:2)
note that almost all ppc binaries are just that... ppc binaries. no optimizations for g3 let alone g5. these binaries are designed run on a 601! so, if you want to get real performance out of yr linux-mac run something like gentoo...
Re:Why did they leave out ... (Score:2)
But I guess if i'd really want to know I could always try out, write myself an article and get slashdotted
In that case, i'd try out Gentoo, perhaps Yellowdog, NetBSD (current has SMP support as far as i know), and benchmark that vs OS X. Perhaps someone out there already have done this, and could enlighten me ?
Re:Why did they leave out ... (Score:4, Informative)
be warned - to get the boosts yr going to have to start at a low stage install and this results in a looong install time. on my imac it took (wait for it) a week to install. and updates take forever too. but, as long as you read up on your use flags and set them well you'll get a fast core system.
Re:Why did they leave out ... (Score:5, Insightful)
Yes, the submission should probably read 'Current x86 Processors Tested with Linux', but in reality the vast majority of individual Linux boxen will be using x86 based hardware so I don't think it's that big of a deal.
But if G5 boxes were shown to be ... (Score:5, Insightful)
But we won't know if we don't look.
Re:But if G5 boxes were shown to be ... (Score:2, Interesting)
Although I've got a feeling that the G5 would do ok but wouldn't end up on top.
I'm not trolling but...
Maximum PC magazine had a review/benchmark in their Jan '04 edition. Featuring an AMD 64 FX-51, P4-EE and a Dual, 2 Ghz G5. (unfortunately the article isn't available online)
To summarize, the G5 finished last overall. The tests even included Photoshop (with filters provided by the folks at M
Re:But if G5 boxes were shown to be ... (Score:2)
Well, you said it yourself. I make small
Re:Why did they leave out ... (Score:5, Informative)
They really should have used a Yellow Dog box or named it an x86 shootout.
Re:Why did they leave out ... (Score:2, Interesting)
I never understood why IBM has stuck with the x86 PeeCees this long. Lord knows I'd like a nice IBM laptop with a Power PC in it. If they could do it for a comprable price.
Re:Why did they leave out ... (Score:2, Informative)
Re:Why did they leave out ... (Score:2)
Yes, but.... (Score:3, Insightful)
Kjella
Re:Why did they leave out ... (Score:5, Interesting)
Having just had my first experience running Linux on something other than an x86, I was curious too.
You can blame it on this very forum [slashdot.org] - after reading the article I bought an Ultra 5 on EBay [ebay.com] and loaded Debian [debian.org] on it last night. Installed most of the packages over my ADSL connection. Worked like a charm.
...laura
Don't use percentage, or use it consistently! (Score:5, Insightful)
but for now most people who buy x86 machines will stick with Windows. so what?
but in reality the vast majority of individual Linux boxen will be using x86 based hardware so I don't think it's that big of a deal.
but in reality the vast majority of individual desktop boxen will be using Microsoft based software so I don't think it's that big of a deal to test with Linux anyway.
Sarcasm apart, I think it's a very bad tradition to consider Linux only as for x86 platform. There are Linux users on other non-x86 platforms. Their percentage is most likely not less than the percentage of Linux/86 users among all x86 users. So, the logic of using any percentage here is basically corrupted.
Linux is multiplatform system. Check the kernel source code for the list of all supported platforms. Kernel - because that wat makes it called Linux, the distributions are usually more platform specific. Also Linux is multi-purpose system - it can be used for servers, for embedded systems and for desktops. The list of oticable desktop systems on the market includes at least x86 and PPC platforms. Therefore considering Linux desktop as only Linux/x86 is not more fair than considering desktop OS only as Microsoft Windows.
Re:Why did they leave out ... (Score:2, Insightful)
Re:Why did they leave out ... (Score:2)
Perhaps Apple marketing paid them not to include the G5's. It would look bad for them if the x86's weren't trounced by the G5's.
Re:Why did they leave out ... (Score:2, Interesting)
Re:Why did they leave out ... (Score:3, Interesting)
Extra Transistors (Score:5, Interesting)
Let's check (Score:4, Funny)
one, two, three...
I'll be back.
Re:Let's check (Score:3, Informative)
A lathe spins an object about a point very fast allowing you to work it into a cylindrical shape.
A planer shaves a layer off the top (or bottom) of an object allowing you to quickly work it down to a desired thickness...
All the same... (Score:3, Funny)
Kjella
Re:Extra Transistors (Score:5, Informative)
I'd actually be suprised if the AMD64 extensions were in this chip...most of what I've been hearing/seeing has Tejas being the earliest we could see them.
Intel and AMD64 (Score:2, Interesting)
In fact, that's probably why Microsoft hasn't released a 'final' AMD64 port of their OS yet. They're expecting a split in the market, and they're not going to get caught in bed with the side that hasn't historically been dominant.
Re:Intel and AMD64 (Score:4, Interesting)
Re:Extra Transistors (Score:2, Insightful)
Example: you have a unit (say, multiplier) that has three stages, each of which can complete within 1 nsec. This means your clock can run at 1GHz (1/nsec). If you divide each of these stages in two then each of these new stages will complete in
Re:Extra Transistors (Score:5, Informative)
Don't forget the extra pipe stages... those count for a lot. I can't say how many transistors, but it should make a dent in the numbers.
A 1 MB L2 cache will be at least 50 million transistors, assuming 6T/bit (1024 * 1024 * 8 * 6), not including sense amps, decoders, tags, coherence, predecode bits, etc.
Re:Extra Transistors (Score:2)
Re:Extra Transistors (Score:2, Informative)
Re:Extra Transistors (Score:2)
Linux becoming a lot more mainstream? (Score:5, Insightful)
Re:Linux becoming a lot more mainstream? (Score:2)
Re:Linux becoming a lot more mainstream? (Score:2)
Re:Linux becoming a lot more mainstream? (Score:2)
Why use Intel anymore? (Score:2, Interesting)
Re:Why use Intel anymore? (Score:2, Interesting)
You know what they say (Score:3, Funny)
Re:You know what they say (Score:3, Funny)
Re:Why use Intel anymore? (Score:4, Informative)
I will always use AMD for my home PC mainly because of the price. IMHO Intel is over priced, you're paying for the name not the product/quality.
Answer: Compilers (Score:5, Insightful)
That's why I still buy Intel.
Re:Answer: Compilers (Score:4, Insightful)
The IAs are roughly comparable timing wise... (Score:2)
Re:Answer: Compilers (Score:3, Informative)
I can't find the link, but I remember hearing once that code optimized for the Pentium actually had a larger speed improvement on the Athlon than it did on the Pentium. I don't recall which models were under test or what their clock speeds were, but it was a few years ago now. It wasn't publicized much, of course, because it bloodied Intel's nose..
The point is that AMD CPUs perform optimized and unoptimized tasks more efficiently (wrt clock speed) than I
MMX came before 3DNow (Score:3, Informative)
AMD licensed MMX and then added on top of it...
This article [informit.com] provides background.
Re:Answer: Compilers (Score:3, Interesting)
On the Portland Group's website [pgroup.com]. If you have the money, they're darn good compilers. Microway [microway.com] sells them as their preferred C/C++ suite, which says something... They support AMD64 too! :-)
The only downside, for some, is that they're Linux-only.
Re:Why use Intel anymore? (Score:2)
But what about AMD? I've personally never seen a commercial touting an Athlon from AMD. AMD should buy some commercial time and tout "Hey we kick Intel's ass AND we're cheaper! Join the rev
Re:Why use Intel anymore? (Score:3, Insightful)
"AMD should buy some commercial time and tout "Hey we kick Intel's ass AND we're cheaper!""
I think part of the reason AMD is cheaper is that they don't spend huge amounts on marketing.
Re:Why use Intel anymore? (Score:2)
Re:Why use Intel anymore? (Score:3, Interesting)
I recently got a 2.8 P4 with HT, I found that with a $150 Mobo the price was very competitive with the latest AMD CPU/Mobo combinations. About the only benchmarks that AMD was b
Re:Why use Intel anymore? (Score:2)
AMD makes motherboards? Also, what does the CPU have to do with the motherboard? Until the PII, Intel and AMD CPUs fit in the same boards. If the board was junk, an Intel chip isn't going to magically re-draw the traces, nor add extra layers to combat EMI.
"Of course way back in the day there were a lot of applications that simply would not work on AMD architecture - this goes back to Windows 9x"
Really? Which ones had probl
Re:Why use Intel anymore? (Score:3, Informative)
The "not 100% compatible" complaint is about ten years old.
My experience, buying a 200 MHz K6 and an 800 MHz Duron, has been good with AMD chips.
The price/performance ratio has always been more attractive for AMD relative to Intel.
I will say that once I had a problem with a K6-III that would lock up after an hour or so - turned out the cheapo CPU cooling fan wasn't moving.
AMD produces fine chips, but their reputation as a "budget" CPU means they get tied into other components of mediocre quality in ord
Re:Why use Intel anymore? (Score:5, Insightful)
I will always use the best priced solution. While I agree that right now that is AMD, Intel is welcome to tempt me -- I can be convinced.
Multimedia editing and encoding, that's what (Score:3, Informative)
Re:Why use Intel anymore? (Score:2)
Then Linux users, who like me pride themselves on the openness of their software, should consider using open standards for their hardware. Such as SPARC [sparc.com], especially the scrummy 64-bit sparcv9. These things can scale up to a 64-CPU box; I'd like to see the Athlon manage that ;-)
Re:Why use Intel anymore? (Score:2)
Re:Why use Intel anymore? (Score:5, Informative)
Good old FUD -- when Windows crashes on an Intel box, it's because of Microsoft. When Windows crashes on a AMD box, it's because the AMD "isn't 100% compatible". Uh huh. I'll buy either AMD or Intel without a hesitation (I'm not a fanboy) where one is the clear winner, but I still see FUD when someone spouts it.
Here's a clue for you -- neither AMD or Intel are x86 processors, but rather both have a converter that converts x86 to their own internal microcode. The x86 standard that they're converting is EXTREMELY well known and standardized (I mean -- other people have to write the software that runs on it). AMD is 100% compatible with core x86.
Intel also has much more R&D into making a chip that won't fry itself if it gets too hot. AMD? Naw, they'd rather make you buy a new chip than bother to make it prevent a problem before it occurs.
Tom, is that you? Firstly, Intel and AMD chips are comparable in heat levels (actually the Prescott sets all new levels for heat generation, making Intel the clear winner in the egg cooker category), but secondly you are correct -- Intel has been more fervent in putting heat protection on their chips, though it is arguably over-engineering. It's like having guidewires "just in case" around all of your high rises. It's generally unnecessary as, apart from Tom's Hardware PR stunts, heatsinks don't fall off processors in regular use. My car doesn't have a special fuel purge if I decide to fill it full of ketchup.
Re:Why use Intel anymore? (Score:5, Insightful)
That's not really a relevant analogy. Something more analogous would be 'my car won't explode if I don't put any coolant in the radiator'. Your car will show the temperature steadily increasing and it may well let you get it hot enough to start a fire or seize the engine. A Pentium engine OTOH would turn itself off after the temperature threshold reached a certain point, thereby saving itself.
What Intel fanboys don't realize is that AMD left it up to the motherboard manufacturers to add proper thermistors beneath the ZIF socket where the cpu goes. While some believe it was irresponsible and some motherboard companies just forgot to add this altogether, most reputable companies have done it for awhile.
You could take issue with the fact that any AMD chip will self-destruct with no heat sink or without adequate cooling, or you could figure you're a total moron for not putting coolant in your radiator and you reap what you sow. Many ways to look at this issue.
Re:Why use Intel anymore? (Score:3)
Experience counts in this one.
still not biting (Score:5, Insightful)
CB
Re:still not biting (Score:5, Funny)
Re:still not biting (Score:3, Interesting)
No, we don't need 64 bit on the desktop. We haven't for the last decade or so, and it won't have that big of an impact immediately.
For gaming? What, you need 64bit color and 64bit sound? No, 24bit is already more colors than the eye can distinguish.. Even if you did, that's the realm of the GPU/APU. For writing emails?
So what if you can linearly address 4TB? A 1 or 2 GB machine is top of the line these days so far as
Re:still not biting (Score:3, Interesting)
Re:still not biting (Score:5, Insightful)
Pretty much, yeah... unfortunate too, since they were such good chips.
For gaming? What, you need 64bit color and 64bit sound?
No, for content creation -- textures, maps, etc. You can't reasonably address >4GB with a 32-bit chip (yes, you can do it. No, it's not fast) and we're reapidly approaching that limit in content creation.
And while 24-bits of color may be adequate, you need more precision than that when doing blends and other operations. Of course, those are largely done by the GPU nowadays, so that's a non-issue as far as the CPU is concerned.
So what if you can linearly address 4TB? A 1 or 2 GB machine is top of the line these days so far as desktop boxes go. The barrier is mainly price, not the addressing capabilities of the CPU.
Actually you can linearly address 256 TB with the current AMD64 chips (48-bit addressing) and 16 exabytes with true 64-bit addressing. That's quibbling though.
The main advantage of a 64-bit CPU (or at least one with direct addressing of >32-bit) is that you can directly map permanent storage to memory. Right now trying to do that is a freaking nightmare and very expensive, since we've long outgrown 4GB of disk space.
And, really, that's not even the main advantage of x86-64. The real advantage is compiling 32-bit code to be aware of x86-64's extra registers, which can lead to a considerable speed up with no other changes.
As far as all the kids running out to spend their allowance on AMD64 chips, that's just them trying to fit in and show how computer savvy they are.
Yeah, by and large. My next PC will probably be x86-64 based though. Why? Because it's not that expensive. You can get a Athlon64 3000+ for just over $200 now. That's only a little more than an AthlonXP 3200+ or a little less than a P4 3.0C. If you're looking in that price range, then why not go for it? You'll get roughly the same price/performance in 32-bit and be able to upgrade to 64-bit when the time comes. If you don't need that kind of performance, that's fine -- save your money. But otherwise it's kinda silly to ignore the potential advantages offered.
Re:still not biting (Score:2, Informative)
One of the reasons to do use the new amd 64 bit chips is that since you'll have to recompile your software for it anyways to use the 64 bit mode efficiently, AMD has been kind enough to add
Re:still not biting (Score:2)
If you are worried about not needing 64bits, it doesn't really matter as the A64 is an all around much better 32-bit chip than A-XP.
-Eyston
BIOS - [Advanced] - FSB - 200MHz - F10 (Score:2)
Apples compared to oranges. (Score:4, Insightful)
I'm sick of hardware sites and the lame "X vs Y showdown" articles. They're utter bullcrap.
Re:Apples compared to oranges. (Score:5, Funny)
Re:Apples compared to oranges. (Score:2)
PPro 200 (Score:5, Interesting)
I run 2 PPro 200 boxen. They are up 24/7/365. I find that linux keeps it lean and the incentive to keep up with the Prescott's just isn't there for me. For a home server I would say keep it simple and reliable. If you want a work engine that will be used for large EDA jobs etc, then the cpu may get you somewhere (coupled with a good system mobo etc). For EDA jobs though we are starting to look at the future wrt 64bit Synopsis builds for the Opteron family.
I am curious if home server users need all this power...
Hedley
Re:PPro 200 (Score:2)
Re:PPro 200 (Score:2)
Sure does! And FreeBSD (my server OS of choice, but not by a huge margin) does also.
However, even my home servers spend more time running applications than kernel code. I run a Zope webserver, LDAP, a decent-size IMAP setup, and a few other services that don't require huge amounts of horsepower averaged over time, but that really make use of a nice CPU over short bursts. I probably only serve 5000 to 10000 web hits per day,
Re:PPro 200 (Score:2)
My main mail/gateway/firewall/webserver is a Dual Celeron 500 and I don't see any reason to upgrade the processors. It ran Red Hat 7.3 very well for years, and now runs SuSE 9.0 just fine.
The only upgrades I've done were disk space related. The last one was a 3ware 7000-6 (IIRC) with 2 120GB Seagates for RAID 1.
Re:Selling? Me too! (Score:3, Funny)
Site's down (Score:2, Redundant)
Re:Site's down (Score:2)
Re:Site's down (Score:2)
Yet Another... (Score:5, Interesting)
Re:Yet Another... (Score:5, Interesting)
Of course. Eliminating a cache miss is huge. Suppose that it takes 300 nanoseconds to get a word from memory.
At 3GHz, that is 900 wasted cycles where each cycle could have potentially retired 3 instructions.
At 6GHz, assuming your memory latency doesn't change, that is 1800 wasted cycles.
It is well known that the memory latency is not keeping up with the clock cycle latency. This is why memory system design is becoming far more important the processor core design, and is also why Itanium SPEC numbers are so good.
Re:Yet Another... (Score:3, Funny)
You make it sound as though cache is the primary factor in CPU performance.
That's... absurd.
Also, the statement is non-sequitur. The Athlon64 3400+ has more L1+L2 cache than the P4 Prescott or the P4 3.2 Northwood, and I believe also the P4EE (Gallatin). The P4EE has 2MB of L3 cache. There was no discernable "winner". There were only a handfull of chips tested. The AthlonFX wasn't amoung them. There were only a handfull of benchmarks. This is
Disappeared (Score:2, Informative)
It gets better (Score:5, Informative)
Those scores are probably not representative of the true performance of the AMD processor but rather of the early stages of optimization of the Linux support for the Nvidia NForce 3-150 chipset. The very same weakness was observed in the past in other reviews that used Specviewperf on Windows platforms, such as this one [tomshardware.com] from THG. Subsequent versions of the Nvidia drivers have since brought noticeable improvements [tomshardware.com].
The AMD scores would likely have been much more competitive if a motherboard based on another Athlon 64 chipset like the VIA K8T800 would've been used for this review.
Re:It gets better (Score:2)
I would have picked a VIA chipset anyday for this test. Although VIA has had issues in the past under linux, these days it's better than the nforce, for maturity and stability if nothing else.
Difference is in price & power consumption (Score:5, Informative)
In all honesty--unless you absolutely need 8GB of memory--there's little difference between these processors in terms of performance. They're all more or less in the same ballpark. Sometimes faster, sometimes slower, depending on the benchmark. None of them is a huge breakout CPU performance-wise.
Here's what's different:
PRICE: There's a lot more than a few percent variation in price.
WATTS: In exchange for your 5-15% speed boost, note that you're getting more than a 15% increase in power usage.
Processor Last (Score:5, Insightful)
Buy brand name components, a decent 7200 or even 10K RPM hard disk, and a GIG of memory. Don't forget input and output either! A cheap monitor, keyboard, or mouse will ruin your experience. Don't skimp on where it counts so you can simply brag about having one processor vs. one that's slightly slower.
Re:Processor Last (Score:2)
I have a simple method for picking CPUs:
Re:Processor Last (Score:2)
Is there a maxed memory test somewhere? (Score:3, Insightful)
I'd love to see a comparison, on linux, Solaris, even windows of Databases crunching big piles of data.
LS
Compare any to any (Score:2, Interesting)
I really don't understand why more sites don't let you compare *any* part to any other, in a database approach, like StorageReview and some online stores. It is in fact relevant to do so for more than interest, you could for example find out if it is more cost effective to use 5 $100 chips or one $1000 chip. But sites do not support this, instead you have to piece it together yourself (if you're lucky and they're using consistent method
No native AMD64 benches? (Score:5, Interesting)
64-bit (Score:4, Insightful)
I was excited to look at these benchmarks because I know that Linux does 64 bits and I really, really want to see what kind of difference there is before buying an Athlon64. But there is no 64-bit testing. What is the point of running on Linux if you don't take advantage of what Linux offers?
Re:nehalem (Score:2)
Perhaps the 'Nehemiah' ? (Score:2)
Lots of info at www.mini-itx.com [mini-itx.com]
Simon.
Re:WHAT MONEY?!? (Score:2)
I guess I should add to my statement... "UNABLE TO TAKE AN F'ING JOKE!!"
Does no one understand sarcasm and humor these days?? Even if the humor isn't very good, it's usually still identifiable to most of us.
GET WITH THE PROGRAM!
Re:WHAT MONEY?!? (Score:3, Insightful)
I'm 32 with short hair, don't drink Mountain Dew because here in Canada it doesn't have caffeine, I live with my girlfriend and 1 year old baby, and I write GPL/BSD code and get paid for it.
But you got it right on the lack of sleep part.
I think *this* is the lowest common denominator of Linux users
Re:WHAT MONEY?!? (Score:2)
I KNEW I hated America Jr. for a good reason!!
Re:WHAT MONEY?!? (Score:3, Funny)
Hey, I am 17, but I do have long hair.
Mountain Dew can marks embedded in their chin
Does Code Red still count as Mountain Dew?
living at home with mommy and daddy
Now this one is just plain wrong. It's a finished basement apartment. At least that is what Mommy calls it.
no job
Now, now, some of us mow lawns in the summer.
never had a date
Who needs a date? I saved up using my lawn job and got a RealDoll(TM) [realdoll.com]!
eyes more bloodshot from lack
Re:WHAT MONEY?!? (Score:2)