Athlon Socket AM2 Review

NerdMaster writes "Hardware Secrets has just posted an Athlon 64 X2 5000+ review, one of the first AMD CPUs to support the new socket AM2. It runs at 2.4 GHz, has two 512 KB L2 memory caches (one for each core) and supports DDR2 memories." However, many are still predicting an end to AMD's dominance in the market thanks to Intel's Conroe.

Comments

[offtopic] 'Print' version is split into 10 pages.

[Whiney Mac Fanboy]

Hardware secrets - you suck.

10 pages not saying very much is (irritating, but) acceptable, but when you split the print article into 10 pages, you've crossed line from greediness to stupidity.

(fires up IE). Oh nice, and there's advertisments on each of the print pages too. How is that supposed to be printable?

Uneven Benchmark

[Anonymous Coward]

According to the last page of the review, they gave the Socket 939 platform 2GB of ram, and the AM2 platform 1GB.

Re:Uneven Benchmark

[moro_666]

even more, is it just me or did they really take a dual core processor and made singlethreaded benchmarks on it and claimed the results with a proud smile in the face ?

if they'd run 2 quake4's at the same time on the machine, the results would be interesting, but all these benchmarks are just quite worthless from the real life point of view. it is nearly identical cpu with just 2 cores of the logic, why do you expect it to differ in any way (most of the single threaded benchmarks are head-to-head, showing off perhaps 1% difference in performance, which clearly shows that the second core was slapping it's sexy hip while the first one was running it's pants off).

Re:Uneven Benchmark

[kjs3]

Apparently, saving the allowance his Mom gives him hasn't added up to the $75 or so needed for another 1GB of DDR2.

Re:Uneven Benchmark

[Jesapoo]

Yeah, I know since I got my two extra arms I've been playing two instances of Q4 lots more.. ;)

Re:Uneven Benchmark

[default luser]

if they'd run 2 quake4's at the same time on the machine, the results would be interesting

That would be true, if not for the fact that Quake 4 is multithreaded.

So are Oblivion and Serious Sam 2.

Re:Uneven Benchmark

[Anonymous Coward]

Four arms?

So what's the sound of three hands typing? ;)

Re:Uneven Benchmark

[Emetophobe]

There is also Quake4 SMP beta patch that improves frame rates significantly if you have a dual core. ~50% improvement with SMP Patch.

Upgrade?

[Enderandrew]

The 939 NForce 4 Ultra/SLI line is showing age so on one hand, I'm glad to see the new tech (and DDR2 support) finally.

However, I don't know that I can convince my wife to let me spend the money on such a large overhaul again. I'm fairly happy with my AMD 64 system at the moment.

Honestly, I just hope AMD maintains their lead long enough for people to start taking notice (like Dell using AMD in the server line).

Re:Upgrade?

[Ihlosi]

The 939 NForce 4 Ultra/SLI line is showing age

Does that mean it has finally reached some degree of maturity ?

And sorry to disappoint you - since the chipset-cpu interface remains the same, the "old" chipsets can be used for AM2 processors, as long as the mainboard has an AM2- and DDR2-sockets.

Re:Upgrade?

[Enderandrew]

NVidia was talking about the NForce 5 chipsets ages ago, and I believe they were holding off for AM2. I'm pretty sure that we'll see new chipsets pretty soon. The main difference with AM2 is the memory controller, which is built into the CPU, so the chipsets really don't have to be different, but there are some things that need improved/fixed from the NForce 4 series (such as the built in firewall, etc).

Re:Upgrade?

[ted.hansson]

NVidia was talking about the NForce 5 chipsets ages ago, and I believe they were holding off for AM2. I'm pretty sure that we'll see new chipsets pretty soon.

According to The Register, that will be later today http://www.reghardware.co.uk/2006/05/23/review_amd _am2_athlon_64_fx-62/

Re:Upgrade?

[Enderandrew]

Now I don't recall where I read it, but about an hour ago I read the breakdown on the 4 different chipsets in the NForce 500 series. Basically there is the uber-overclocker's SLI edition, the silent SLI edition, the baseline series, and the Crossfire series. Here you go. I think this was the article. http://www.nforcershq.com/article6287.html

Re:Upgrade?

[stupidfoo]

The 939 NForce 4 Ultra/SLI line is showing age

How so? My current setup (ASUS A8N-SLI Deluxe, 3800+, 7800GTX, 1 GB DDR) has no problems running any of the most intensive games and/or apps. About the time that Vista comes out I'll probably just upgrade the processor to a 4800+ (which will be about $300 by that time), get a second 7800GTX, and slap in another 1 GB of DDR. All for about $600. And then I should be good to go for another year or two.

Re:Upgrade?

[Enderandrew]

The chipset itself is showing age, not necessarily your PC.

Early SLI offerings leave much to be desired in efficency and design. Furthermore, the NForce 4 series was innovative in features, but since so many of these features were new, they were buggy or weren't implemented horribly well.

We're due for a more stable NForce 5 series.

Re:Upgrade?

[Anonymous Coward]

The 939 NForce 4 Ultra/SLI line is showing age

I'm fairly happy with my AMD 64 system

So which is it? Either your 939 system is running slow or it isn't. Unless you live in backwards land where slow makes you happy. And furthermore, be a man and tell your wife what you will and won't buy. You're a perfect example of the emasculation of modern society. This is how you should run your life: Do what you want, let her burn the meatloaf, and keep playing Oblivion at 1280x1024 on your AMD 64 system.

Re:Upgrade?

[Bender]

The 939 NForce 4 Ultra/SLI line is showing age so on one hand, I'm glad to see the new tech (and DDR2 support) finally.

I don't understand. How is the nforce4 showing its age? Performance is excellent, and as reviews have shown, better than the AM2 platform right now.

Re:Upgrade?

[Anonymous Coward]

Honestly, I just hope AMD maintains their lead long enough for people to start taking notice

Who cares? I'm going to use whatever is the cheapest/fastest/best platform. I don't care if it's AMD or Intel. What are you a fanboy?

I'm currently using an Opteron system but will probably switch to Intel/Conroe when there are fast enough setups. I'm sick of the crap-ass socket-940 motherboards. AMD should make their own boards and full chipsets because the 3rd-party stuff is garbage. Intel does it better (disk controllers and such) but the Prescott processors and everything based off those cores suck donkey balls.

Re:Upgrade?

[stupidfoo]

True. Good points. I'll especially agree with you on the buggy part. I had a LanParty board before the ASUS. It was the most unstable piece of crap I've ever owned. Tons of features and a really sexy piece of hardware, but it was just worthless.

Maybe I'm just trying to convince myself that I don't need a new system because I know that if I were to purchase one my wife would remove a very important body part.

Oh well... at least I convinced her that we should get a Nintendo Wii. Although all I had to tell her is that she could play Super Mario Brothers and Dr. Mario on it.

Re:Upgrade?

[clem]

Maybe I'm just trying to convince myself that I don't need a new system because I know that if I were to purchase one my wife would remove a very important body part.

Your left mouse-clicking finger?

Re:Upgrade?

[complete loony]

Tell me about it, using the internal NIC gives me file corruption on ftp transfers, and no I haven't found out why yet. Most of the time I connect to the net using PPPoE with no corruption, and I assume using less features of the chipset. But if I connect more normally or use a local ftp server, some download will fail almost every time.

Sigh.

2.6 Ghz, not 2.4Ghz

[adam1101]

One of the first sentences of TFA states that the 5000+ is clocked at 2.6Ghz.

Ignore these benchmarks

[Enderandrew]

They actually tested the 939 rigs with 2 gigs of memory, and the new rig with 1 gig of memory. They said they didn't have 2 sticks of DDR2 for the AM2 rig, but then they should have only used 1 stick in the 939 rig.

When benchmarking, you should try to keep all test systems as comparable as possible. I really am disappointed by what I consider a glaring oversight.

Seriously, for shame.

Re:Ignore these benchmarks

[Anonymous Coward]

Yes, it showed AM2 in a favorable light. Two RAM sticks increase overall timings. None of the benches they did would exceed over 1GB memory required, so there would be no HDD swapping involved. In other words, expect a ~1% drop relative to their numbers if they add an extra RAM stick in the AM2 machine.

Either way, it's statistically insignificant. Don't get your panties in a wad over it.

Re:Ignore these benchmarks

[Barny]

The testing of one rig with 2 sticks of 1G ddr ram (2x64bit bus, accessed in dual channel mode so 128bit) is VERY much inacurate against a rig running one stick of 1G (1x 64bit bus, can only be accessed in single channel, 64bit, mode), and I believe that is a serius error.

However, AMD is going to lag a little behind in the short term, although the memory latancy is still allmost half of what intels current parts are capable of (check out theinquirer.org for some interesting insights into thise issue).

Re:Ignore these benchmarks

[Enderandrew]

The thing that really got me was the conclusion page. For many people, that is all they read. In said conclusion they said the DDR2 support wasn't necessary that great because the FX-60 beat out the 5000+ in several test.

In all the CPU specific test, the 5000+ came out on top. In tests that benchmarked total system power, the FX-60 rig with 2 gigs running in dual channel mode came out on top.

If these guys don't know how to put together proper testing rigs, let alone extrapolate results, then AMD shouldn't be wasting test equipment on them. They can send processors to me, and I'll put up some serious benchmarks, including detailed overclocking, burn-in testing, temperature readings, etc.

Re:Ignore these benchmarks

[Anonymous Coward]

In said conclusion they said the DDR2 support wasn't necessary that great because the FX-60 beat out the 5000+ in several test.

Huh? Second paragraph:

But, on the other hand, we have to consider the future and what is behind AMD's strategy in going to AM2 socket and DDR2 support. In our opinion, what is important isn't the release of Athlon 64 5000+, but the possibility of increasing the memory bus clock rate in the near future. For socket 939 platform this was not possible since the top DDR official standard was DDR400/PC3200. With AM2 AMD can finally go beyond that, as DDR2 official standards include at least three speed grades above that: DDR2-533, DDR2-667 and DDR2-800. That is promising.

Re:Ignore these benchmarks

[Joe The Dragon]

also the 5000 has only 512k L2 pre core where as the FX-60 has 1meg pre core.

Re:Ignore these benchmarks

[MadRocketScientist]

From TFA, Page 3: Memory: Two Corsair CM2X512-8500 modules with 512 MB each, installed on DDR2 Dual Channel configuration (using 4-4-4-12 timings) for socket AM2 CPUs.

Re:Ignore these benchmarks

[Enderandrew]

However, in every pure CPU-specific benchmark the 5000+ beat the FX-60 (not by much, but it did).

Re:Ignore these benchmarks

[valshaq]

But with only one stick you couldn't profit of the dual channel memory controller. This sucks.

Re:Ignore these benchmarks

[Anonymous Coward]

Here is a link to a more complete review:
http://www.short-media.com/review.php?r=316

Re:Ignore these benchmarks

[ocbwilg]

But with only one stick you couldn't profit of the dual channel memory controller. This sucks.

It doesn't suck. Any other hardware review site would have had another DDR2 stick to put in, or would have at least gone out and bought one. The reviewer sucks.

Re:Ignore these benchmarks

[Barny]

Woops, missed that one, hehe, chalk one up for crappy 10 page layout :/

One thing they do however neglect to point out, after their wonderfull "better buy than Athlon 64 FX-60 if you are really a performance freak but want to save some money.", is that the fx-60 is is unlocked so far as multiplyer goes, that is the BIGGEST draw card for the fx range, and primarily why you would buy one, and no performance freak should ever look at their bank ballance, its bad for your heart ;)

Re:[offtopic] 'Print' version is split into 10 pag

[Shadows]

Summary: Blah blah, blah, blah blah blah... and finally something interesting:

[from page 10]

We didn't see any performance improvement by the use of DDR2 memories instead of DDR. In fact, Athlon 64 FX-60 was faster in several situations, showing that at least for the software we used having a bigger L2 memory cache is better than having DDR2 memory instead of regular DDR.

But, on the other hand, we have to consider the future and what is behind AMD's strategy in going to AM2 socket and DDR2 support. In our opinion, what is important isn't the release of Athlon 64 5000+, but the possibility of increasing the memory bus clock rate in the near future. For socket 939 platform this was not possible since the top DDR official standard was DDR400/PC3200. With AM2 AMD can finally go beyond that, as DDR2 official standards include at least three speed grades above that: DDR2-533, DDR2-667 and DDR2-800. That is promising.

Welcome, Intel

[eebra82]

I for one welcome Intel to the top performance game. I used to be all happy about AMD gradually taking over the market ever since their first Athlon slot CPU:s. This has also caused the pricing war significantly since AMD already knew they kicked Intel in most parts of all reviews.

Now that Intel is back, we can finally see some heavy competition between the two. The Core Duo is a superb processor and I am eagerly awaiting my MacBook to arrive and I can't wait to see the second release of the Core Duo.

Remember what it was like a few years ago? I used to follow the price charts of CPU:s for drops and they were a lot more frequent than they are today. So now it's easy to say that we should get the same competition all over again and I am quite sure that Hector Ruiz at AMD has a backup plan ready to be enrolled this year.

So once again, welcome, Intel!

Re:Welcome, Intel

[Whiney Mac Fanboy]

So once again, welcome, Intel!

Intel might be the king in the performance-per-watt race, but AMD is still the king in the far more important performance-per-dollar race.

It's a pity Apple didn't go AMD as well as Intel for their supplier of x86 chips - 'cause I always feel like I'm overpaying for an intel product :-/

Re:Welcome, Intel

[chrismcdirty]

You feel like you're overpaying for Intel, but you didn't ever feel like you were overpaying for PowerPC in the past? I have a feeling that if Apple had used AMD chips, you'd still be paying the same price, but more of it would go directly into Apple's pockets.

Re:Welcome, Intel

[Whiney Mac Fanboy]

You feel like you're overpaying for Intel, but you didn't ever feel like you were overpaying for PowerPC in the past? I have a feeling that if Apple had used AMD chips, you'd still be paying the same price, but more of it would go directly into Apple's pockets.

I did feel like I was overpaying for ppc - but I guess you're right. Apple products are just overpriced.

I guess I'll just stay happy with my very cheap AMD laptop :-)

Re:Welcome, Intel

[spectrumCoder]

Could you post a link to the CPU price charts you're referring to? When timing the purchase of a new CPU (or any PC component for that matter) it'd be great to be able to buy when the price has bottomed out rather than when it's in the middle of a steep decline.

Re:Welcome, Intel

[eebra82]

There are quite a few of them out there. Here's one for CPU:

http://www.sharkyextreme.com/guides/WCPG/article.p hp/3607116

And here's one for RAM:

http://www.sharkyextreme.com/guides/WMPG/article.p hp/3607991

Re:Welcome, Intel

[GoatMonkey2112]

I somewhat agree, but Intel with no competition is no better at lowering prices. AMD needs to stay close and we'll be set for some price drops.

Re:Welcome, Intel

[NutscrapeSucks]

You feel like you're overpaying for Intel, but you didn't ever feel like you were overpaying for PowerPC in the past?

Speculation was that Apple paid under $50 for a G4 CPU.

For a large OEM like Dell, I doubt there's any huge difference between Intel and AMD pricing. Apple however is pretty much only using the luxury Core Duo parts, so they are probably spending a lot more money on CPUs and saving it elsewhere by using Intel chipsets and integrated video. If component costs were really a huge concern for Apple, they'd be shipping boxes with Celerons and Pentium-Ds. But with 20-30% profit margins, who cares?

Re:Welcome, Intel

[spectrumCoder]

I'm aware of price watches, which give the week-on-week price changes for components, but what I believe would be really handy is a graph of the price of any given component over time (with a time range of anything from three months to three years).

One could then analyse the steepness of the curve and come to some conclusions, such as the best time to buy a graphics card is nine or twelve months after release, LCD panel prices are unlikely to experience the same rate of decrease in the coming twelve months than in the previous twelve months, etc. (these statements are just examples of the possible sorts of conclusions)

I'm guessing that CPUs have a reasonably shallow curve (since the product lifespan is longer than the typical motherboard or graphics card), but I'm not sure.

Re:Welcome, Intel

[alep85]

I think you couldn't be further from the truth in terms of Apple's decision. AMD's Turion and mobile Athlon chips are CRAP, they have poor battery performance and they can't come close to matching the Core Duos in performance. Perhaps you should realize that Apple is going for performance in a small form factor. AMD just doesn't have good mobile chips to do that. For the PowerMac replacement, yes, AMD would be a great choice, but with Conroe and Woodcrest coming soon, I think we'll begin to see Apple's choice with Intel was a good one.

Re:Welcome, Intel

[Anonymous Coward]

Intel might be the king in the performance-per-watt race, but AMD is still the king in the far more important performance-per-dollar race.

If you look at what Conroe is set to be priced at, it will be half the price with better performance than the closest AMD parts (as they are today anyway). All I've seen from AMD since they took the performance crown is just as expensive as the Intel of old. AMD is looking to be in a world of hurt.

Of course, all this means is that you always need competition around, regardless of who's ahead.

Re:Welcome, Intel

[Bert64]

Turion was faster than the centrino chips, which were it's closest competitor at the time...
Infact, Turion is 64bit, in many benchmarks faster than centrino, and depending on usage patterns, could consume less power.

On the other hand, intel have dual core laptop chips and AMD dont yet, tho turion is still 64bit and intel has yet to release a 64bit laptop chip.

Re:Welcome, Intel

[default luser]

but what I believe would be really handy is a graph of the price of any given component over time (with a time range of anything from three months to three years).


Happy Birthday!

Anandtech's Real-Time Pricing Engine gives you real-time prices, plus a graph of the last 6 months of price changes.

I'm amazed at the fact that they log 6 months of history, and offer it to the public at no charge. Most accounting firms charge through the nose for this kind of pricing trend data.

I'm guessing that CPUs have a reasonably shallow curve (since the product lifespan is longer than the typical motherboard or graphics card), but I'm not sure

In recent years, you'd be correct. The x86 CPU market has come up against a wall, because you can only parallelize scalar x86 code so much on-the-fly. Improvements in CPU architectures like Conroe now take MUCH longer to design, for less and less performace gains than in previous years. Also, the percentage megahertz gain from process improvements is also hitting a wall, which is another reason CPU prices remain stagnant.

Re:Welcome, Intel

[WuphonsReach]

Most items over at www.pricescan.com have price trend charts. It's not a chart for the overall category, just charts for each individual item.

(I've been using PriceScan for quite a few years... maybe as far back as 1998? I still end up ordering from places like NewEgg or MWave.com or TheNerds.net but it's good for seeing prices.)

HotHardware's Coverage Of AM2, Much More Detail

[Anonymous Coward]

The folks over at HotHardware.com also have a very detailed performance evaluation of AM2 Athlon 64, right here

Problems with memory dividers

[Delph1]

What many seem to have neglected is that there is an evident problem with odd CPU multipliers. AMD has no support for "half" memory multipliers (4.5/5.5/6.5/etc), which means that you will actually not be able to run memories at their full potential when using processors with odd multipliers (7/9/11/etc):

http://www.nordichardware.com/Reviews/?page=3&skri velse=481

Re:Problems with memory dividers

[Anonymous Coward]

Or you know, you can solve the problem by overclocking the all of 200 MHz that you need to get the higher multiplier. I mean, most these CPUs tend to OC for a bit more then just 200MHz anyways.

Re:Intel's roadmap to the Cornroe and beyond

[Anonymous Coward]

Yea, Cornroe is a total joke. Check out the Hexus Review. Or maybe not.

Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

My 1.66GHz Athlon XP 2000+ does the job well and is still very often limited by memory (1GB) and harddisk bottlenecks. That's a 3 year old system, and CPU+Board cost only $140 back then. Am I really expected to pay that price several times over to replace a reliably working system and get less than a 2x performance increase?

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

How did you compute the speed increase? A stock Althlon XP 2000+ doesn't even take advantage of PC2700 memory speeds. I think its a safe bet that you would benefit from a serious speed increase on a new motherboard + processor + ram + hdd. Why? Newer ram is very fast although it does have more latency. Its possible to buy 10k sata or 15k scsi disks now which can feed more to the cpu. You are right that your system and any new system you buy is still slow due to disk io. That will always be the case.

I think what you meant to say was that no new software has come out which will force you to upgrade. Windows Vista is not out. Many games can still run on P4 2ghz chips. This reminds me of my first computer purcahse, a pentium @ 100mhz. It ran 3.1 apps fast as hell and then Microsoft released 95 which forced all my friends to buy pentium systems.

You are right that you should not buy a new system right now. Don't discredit the speed increase though. You just don't need it.

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

Unfortunately comparisons of AthlonXP CPUs with current models are hard to find, so I have to derive my estimates from marketing: I suppose that the speed ratings of AMD CPUs are still meant to say "about as fast as an Intel Pentium IV line processor clocked at" (Sempron~Celeron, X2~DualCore). The higher end processors are still somewhere around 4000. A processor which costs what mine cost 3 years ago is rated in the vicinity of 3000. That's a 1.5 increase in 3 years, and just for the CPU. Because many tasks are still limited by other factors, I expect a system which is about 30% faster overall than my current one. Or maybe I'm talking out of my ass and AMD needs to get some relevant comparisons to the people, like "this is your system, this is what's on the market today". Who cares which processor is 3% faster for another 100 bucks?

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

Very much out of your ass...1) You are ignoring the memory increase you are going to get. There are performance differences between the memory bus on the XP2200 and on the Athlon 64 (or the X2). You also bought your processor near the death of the XP line. I had my first Athlon 64 almost two years ago, on the now mostly dead 754. Remember, the cores have changed a few times over as well.

Now go here and compare the T-Bird 1400 to just about any chip in the chart. You notice that the AMD64s near 3000+ are all almost twice as fast in divx and performance is hardly close in most tests. Now granted the 2200+ is a bit faster then a 1400, so go here and use the charts to get an idea of the difference between the t-bird 1400 and the XP 2200+. Oh and remember, that 1400 is actually the speed on the T-bird. The 2200+ is only running at about 1800 unless overclocked.

Overall, you should be seeing anywhere from 50% on up in performance and that is just by fudging the numbers together a bit. In an actual test, you might find the improvement to be much better. Also note, some of the Sempron's could provide smaller improvements in performance for some things, so you could possibly get a cheaper CPU and not go with a 64. Prices on 754 should start falling more soon, I would think. With AM2 and Socket 939, I just cannot see AMD supporting more then two sockets at once. Socket A is already mostly dead and 754 has been falling off slowly. But just remember that would mean moving from one dead socket to another.

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

There are many tests where the difference isn't quite as clear, so I suspect that Divx is highly optimized for SSE, which the Athlon 1400 didn't support. All in all it looks like the 1.5 estimate for just the CPU at the same price point wasn't too far off. That's definitely not enough to throw out a working system. Most jobs that are too much for realtime usually require a longer wait anyway, so it doesn't matter much if the processor is 1.5 times faster or not. The Athlon XP and Sempron speed ratings appear to be roughly comparable.

Re:Buying PCs isn't as exciting as it used to be

[LurkerXXX]

Its possible to buy 10k sata or 15k scsi disks now which can feed more to the cpu.

They had 15k scsi disks back when he bought his rig as well. I've got some next to my desk here which have been running for more than 3 years. Nothing new here. Just a bit cheaper.

Re:Buying PCs isn't as exciting as it used to be

[Kyro]

i love my k7 xp2000+ :D very speedy.
it also crunches dnetc faster than a single 2.8 GHz XEON IBM server i have at work.

Re:Buying PCs isn't as exciting as it used to be

[ArcherB]

My 1.66GHz Athlon XP 2000+ does the job well and is still very often limited by memory (1GB) and harddisk bottlenecks. That's a 3 year old system, and CPU+Board cost only $140 back then. Am I really expected to pay that price several times over to replace a reliably working system and get less than a 2x performance increase?

Where do you get 2x performance? Are you just looking at the clock speed? I ask because my A643000+ is a good clip faster than my AXP3200, both at the same clock speed.

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

I agree, to add another anecdote, my 6 year old Athlon 700mhz does the job just fine. Gaming is 100% out, but everything else is fine, including some development and graphic design work. 90% of the time, I am really just using IM or surfing the web though anyway...
-K

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

"My 1.66GHz Athlon XP 2000+ does the job well and is still very often limited by memory (1GB) and harddisk bottlenecks. That's a 3 year old system, and CPU+Board cost only $140 back then. Am I really expected to pay that price several times over to replace a reliably working system and get less than a 2x performance increase?"

Yes. That is, unless you just run an email box, then... still yes. You'll need the additional horsepower to more enjoyably Shoot The Duck while checking your AOL mail.

Re:Buying PCs isn't as exciting as it used to be

[ZERO1ZERO]

http://tomshardware.co.uk/2005/11/21/the_mother_of _all_cpu_charts_2005uk/page59.html This chart gives a good indication of expected speed. You are right as well in that as well as CPU speed increase the overall system will increase in speed due to newer memory, and I/O etc. Also, i think the Athlon speed ratings are based on a rating of 1000 being derived from the orginal Athlon 1 GHz processor. I don't know if this still applies today, (i thought it did). The previous poster's method of assuming that the XP+ figures are relative to the Intel chips is misguided. Shoot me if I'm talking bollocks.

Re:Buying PCs isn't as exciting as it used to be

[Burning1]

As much as car anologies fall flat on their buttux, I think this one is reasonable:

If you want to go 100MPH you can buy a Civic for $15K.
If you want to go 200MPH you can buy a Zonda for $300K.
If you want to go 250MPH you can buy a Bugatti for $1 Million.

Sometimes I think computers are spoiling people. Small improvments can be very expensive.

For what it's worth though, you don't have to upgrade and you aren't expected to upgrade every time AMD makes a 20% bump in speed. New motherboards are for people who need a new system, not for people who want to make a smaller speed improvment.

The reviews are good for the guy with a old system like mine. "Is it worth spending an extra $50 for the latest and greatest, or can I go with the previous generation?"

Re:Buying PCs isn't as exciting as it used to be

[Anonymous Coward]

The thing is, I feel like I own an old system. AGP graphics card are being phased out now. That's a new experience to me. Usually the main system fell behind enough to warrant a new CPU/mainboard/RAM combination before the other components became scarce or unavailable. I guess what I'm trying to say is that processor speed has stopped being the primary reason for upgrades. These days it seems as if a whole optical storage standard can pass through before you need more processing power.

Re:Moore's law

[Anonymous Coward]

For the umpteenth time,
http://en.wikipedia.org/wiki/Moores_law

Yes it does, because it deals with transistors counts, not cpu cycles.

Re:Intel's roadmap to the Cornroe and beyond

[Vlad_the_Inhaler]

For the last couple of years, AMD products have been the best in so many ways and Intel have been producing heating units. Now the Empire has Struck Back - Dell will be very happy about that.

I personally have absolutely no interest in 'the fastest thing in town', gamers may care but Cool, Quiet and Reasonably Fast are easily enough for me.

From what I hear, the Conroe is slightly more energy-efficient than AMD's offerings.
How does it scale? The Opteron is designed to be a server processor more than anything else.

HardOCP review, link to print version

[Shivetya]

http://enthusiast.hardocp.com/articleprint.html?ar t=MTA2NQ==

Much better than the posted story which is nothing more than an advertisement for Hardware secrets.

When will /. editors review entries to prevent this abuse? If anything, when new hardware is released we all know multiple sites will cover the release. That means, put links to the more popular review sites into the story instead of helping one person get his ad hits.

Much More benchmarks are available on the web

[Anonymous Coward]

Small selection:
- HardOCP
- X-Bit Labs
- HotHardware

up to date list here

Re:Much More benchmarks are available on the web

[amcdiarmid]

Conclusion page at the register. http://www.reghardware.co.uk/2006/05/23/review_amd _am2_athlon_64_fx-62/page7.html

Verdict

These are all some pretty impressive numbers, but bear in mind that we tested the AM2 platform with the FX-62 - other AM2 processors won't show any big improvements over their Socket 939 counterparts. It's a real shame that AMD didn't manage to squeeze some more performance out of the new CPUs, but this will hopefully come with time. For now, the AM2 processors aren't worth the investment if you have a Socket 939 system, unless you want the fastest CPU out there, the FX-62. However, this might not be the case for long if the Core 2 Duo lives up to Intel's hype.

-- Look out: attributed block cut/paste;)

Re:Much More benchmarks are available on the web

[mikemuch]

Also http://www.extremetech.com/article2/0,1697,1966062 ,00.asp which tests the FX-62, too, and uses even testbeds.

am2 seems to be doing fine

[mAriuZ]

Here is another review of the new socket/ddr2 amd part
It's great to look at fx-62 results - it looks like only that processor (or if you overclock it) can
use the available bandwidth
        "Frankly speaking, it's the main competitor who must be bustling now. AMD is doing great anyway. At least in terms of CPU performance. Durability of the K8 core and its capacity to adapt to new market realia is admirable: having lived without major modifications through two process technologies, dual cores, and now a new memory controller, this core meticulously responds to each improvement with performance gains. We were very skeptic about future chances of the new AMD platform against the new processor core from Intel (Intel designed the new core nearly from scratch, while AMD K8 is rather old), but our tests warmed up our interest. The situation may turn out not that simple"

AMD Catches Up in Technology and Shoots Out in Performance
http://www.digit-life.com/articles2/cpu/amd-athlon -64-fx-62.html/

Re:am2 seems to be doing fine

[Bert64]

There you have the difference between Intel and other processor designing companies...
Whereas Intel will design a processor which is barely adequate for today, with no thought for future expansion, and then either kludge the existing design in horrible ways, or have to throw away the whole steaming pile and start again...
Other companies think about the future, when AMD designed the K8 they already had dual core and faster memory controllers in mind.

wishful thinking

[sl4shd0rk]

AMD has had the edge for years, there was just a lot of denial and ignorance of it. People stuck with Intel simply because nobody wanted to spend a couple hundred bucks on a processor and end up being the guinea pigs for AMD.

Re:wishful thinking

[LaughingCoder]

Not me. I've had an "Intel Outside" sticker on my front door for over 2 years now!

Re:wishful thinking

[GiMP]

"Guinea pigs" is right. I've been a long-time user of AMD products and I haven't been happy. The problems have been less with the chips than with the chipsets, but what good is a processor without a quality motherboard? I've built a number of systems with AMD processors and from issues with overheating, to broken usb controllers that would reboot my box if I plugged anything into it, it is fair to say I've had my share of problems. I've used chipsets from VIA, Nvidia, and AMD.

On the Intel side of the pool, I've found that Intel's chipsets are the absolute top-notch, best quality out there, works as promised. I've never had a problem with an Intel motherboard or one based on their chipsets.

What motherboard should one use for their AMD system? I don't think a single AMD-compatable chipset manufacturer has consistently released quality chipsets. VIA is probably the best bet, which I don't believe is saying a whole lot.

As far as performance-per-watt is concerned, for those saying that doesn't matter, think again! This can be a significant difference for enterprise customers. In the datacenter, less-power-consuming and less-heat-producing processors are higly desirable. Electricity isn't free, not only does lower wattage have a direct power savings, it also saves power on the AC units. Furthermore, more power can mean additional costs in both physical space and per-port costs with switched PDU units. This is compounded when using dual power-supplies!

Finally, "AMD is cheaper" is a myth in the current market place. Dell doesn't do AMD yet. If you want to buy pre-built machines off the assembly line, you can choose between an AMD machine from HP/Compaq, IBM, or Sun.. or a machine from Dell. Guess which will be the cheapest? Guess which will have the best warranty? It won't be any of the AMD machines.

Re:wishful thinking

[GiMP]

Before I get flamed, yes, I've heard that AMD has been doing better with their heat than Intel in recent times. If anything, that section of my argument isn't specifically geared towards either camp.

That said, chipsets are more important than heat. As expensive as it is to power machines and to keep them cool, it is even more expensive to have a machine melt into a klump of toxins.

Re:wishful thinking

[Aladrin]

I have always bought Intel because of the stability. I never have to wait for an app or game to release a patch to fix the AMD issues. (Not that there's actually been a problem with that for a while, though.)

But 'AMD Cheaper' is NOT a myth. You can't say 'X isn't true because in Y area it isn't true.'

You are only looking at major computer manufacturers, and then, only a few of them.

Build your own rig. AMD is quite a bit cheaper than Intel if you do.

Major manufacturers have many political and legal hoops to jump through, and that skews their prices crazily. (If you don't believe that, research why computers with Windows installed are the same price as ones with a free linux distro installed.) You can't base your entire analysis of cost on how they price their computers.

Instead, price the components individually or check prices at a 'Mom and Pop' computer shop. You'll see that AMD is consistantly lower in price. (You'll also see that they still manage to sell the higher priced Intel machines as well.)

Re:wishful thinking

[GiMP]

> But 'AMD Cheaper' is NOT a myth. You can't say 'X isn't true because
> in Y area it isn't true.'

Clearly, I meant it is a myth within the server space, and possibly in the corporate workstation market as well. I can't even get 1U AMD servers from a certain "whitebox builder" at a cheaper price than intel machines, as they claim that they have to pay more money to get a motherboard that won't melt.

However, this isn't to say that it isn't a myth in the desktop market too. Whitebox has problems. A truely unique, untested configuration has many possible problems and unexpected side effects. White-box systems are NOT on par with OEMs in terms of quality, they simply aren't. Will you get Doom and Half-life to run? Sure, but would you trust your company's well-being to components from various manufacturers pieced together in a truely unique configuration? Or, perhaps, would you prefer to trust a tested and researched configuration used by thousands of customers?

I'll take it one step back, however, would you trust your marriage on a whitebox computer? My wife is tired of whitebox and is very unhappy when it fails, because computers do fail occassionally, and more often, my experiences have been that whitebox fails more frequently than OEM. Do I hear echos of "untested" and "unique"? YES! Often many problems arise from an overloaded the PSU, cheap PSU, cheap ram, cheap motherboard. I believe that if you want to build a system that is as rock-solid as an OEM, you really do need to pay OEM prices, and I don't mean store-front costs, I mean their R&D costs. A quality OEM won't just sell you the first thing that boots, they actually test machines and will swap out parts that don't work right. Do you do that, or do you just say, "crap, the USB controller doesn't work! I'll have to add an external card!"

> You can't base your entire analysis of cost on how they price
> their computers.

Certainly if my manufacturers of choice are big OEM shops, then yes, I can base my analysis how they price their computers! Mom and Pop shops aren't where the majority of US consumers shop at, and in my experience, those shops tend to overcharge customers anyway.

There isn't any logic to the statment: "I'm buying an AMD system from HP at $1000 more than I'd pay to Dell for an Intel system, but it is an AMD.. so I must be saving money. Although I have only 1/3 of the warranty, I'm sure I won't need that with AMD."

Re:wishful thinking

[tomstdenis]

Tool.

I've had nothing but success using nForce4 and nForce4 Pro based motherboards.

On the other hand my i915G motherboard for my 775 pin Prescott can't access the CDROM in Linux [reliably anyways]. My Gigabyte i945 was fine though.

Stop buying crap ECS boards or low end KT or nF series. Spend the $100 and get a nF4 ASUS or Gigabyte and get over it.

Tom

Re:wishful thinking

[GiMP]

Gigabyte.. I've bought two of their motherboards. The first had an AMD chipset and would reboot upon plugging in a USB device. No fix from Gigabyte was ever provided. I believe that a Windows driver fix was made available, which did something silly like disable the +5V line. It never worked on Linux. I ended up buying an add-on PCI card for USB.

The other gigabyte motherboard I bought was crashing at high loads, even with low temperatures. The jury is still out on that one, as it could still possibly be another component, but my tests showed the disk, memory, and power supply to be fine.

nVidia chipsets. I admit they're not too bad, but I wouldn't recommend them for a server. For a workstation, they're not too bad, I've used them. The Linux drivers could use some improvement. The onboard components tend to be pretty top notch with nVidia, as far as onboard components are concerned. If you don't have need for onboard components, nVidia chipsets are overkill.

Re:wishful thinking

[ponos]

I've been a long-time user of AMD products and I haven't been happy. The problems have been less with the chips than with the chipsets, but what good is a processor without a quality motherboard?

It would be accurate to say that the first K7 had some lousy chipsets. Then again, I own one and it routinely reaches 40+ days of uptime (note that this is an el-cheapo box that I use as proxy server). The newer Athlon XPs had decent chipsets, with some exceptions. You may have suffered with some of them. The Athlon64 chipsets are objectively quite good. There is a selection of ATI, NVidia and VIA to chose from and all major manufacturers make overclocker grade motherboards with top components. Especially now that the memory controller is on-die, the chipset is not as important as it used to be. In conclusion, I don't think you can really say that the new AMD processors don't have quality motherboards and chipsets.

As far as performance-per-watt is concerned, for those saying that doesn't matter, think again! This can be a significant difference for enterprise customers. In the datacenter, less-power-consuming and less-heat-producing processors are higly desirable.

Performance per watt in the AMD side has been MUCH better until intel made their new, shiny 0.65nm Core and suddenly "invented" this concept. It used to be GHz when Intel used the NetBurst architecture, now it is performance per watt. Note that AMD was the first to supply power saving for server and desktop processors. All in all, the difference between Athlon64 and Pentium 4 was MUCH bigger than the advantage of new Intel processors over Athlon64. Even the new Athlon64 AM2 3800+ dual cores are rated at 65W (and 89W) versions.

Finally, "AMD is cheaper" is a myth in the current market place. Dell doesn't do AMD yet. If you want to buy pre-built machines off the assembly line, you can choose between an AMD machine from HP/Compaq, IBM, or Sun.. or a machine from Dell. Guess which will be the cheapest? Guess which will have the best warranty? It won't be any of the AMD machines.

I have to agree with you here. Not because of Dell (Dell is a minor player in my country) but because the prices for cheap dual core AMD processors are much higher than the cheap Pentium dual cores. If you want dual core, AMD will make you pay. If, on the other hand, you are looking for REAL cheap, Sempron is a great way to go. I repaired my mother's PC with a Sempron 64 3000+ (with Cool and quiet) plus an Asus motherboard and this thing is amazing. Costed approximately 130E WITH motherboard and runs extremely cool with the stock cooler (idles at 991 MHz, 1.1v).

P.

Re:wishful thinking

[jp10558]

I don't know where you buy consumer boxes (can't speak to coporate servers), but I've yet to find an OEM box that isn't crap. Crap packed with cheap propritary parts, odd designs, sometimes custom PSUs, occasionally the CPU SOTTERED to the motherboard etc...

The very best OEM I've ever dealt with was MicronPC, and they went out of business I think.

Of course, neither one of us is statistically relevant, but I've had the exactly opposite experiance of you.

Re:wishful thinking

[Anonymous Coward]

Are you fucking serious? You must have shit whitebox builders or amazing luck with OEMs. In a perfect world I would hire a gang of highschool kids and just run my own whitebox assembly on-site - that way I get the cost savings and complete control over components. There is a huge overclocking community which pushes the components to the edge, if you don't want to take the time to read their forums and figure out what gear is the most stable then you can hire one of them to figure it out for you - it's not hard! You don't need to do full on R&D - it's already been done for you by thousands of people testing thousands of hardware configurations in extreme circumstances. I've built numerous whiteboxes myself, I always do the research ahead of time and have never had a single issue. It's not hard to build a rock-solid box.
Instead of going the on-site assembled whitebox route I'm saddled with Gateway workstations and Dell servers. The servers haven't had too many issues, more than I'd like but Dell is usually not too bad about dealing with it. The Gateways though - we have a pretty large volume coming through and they are constantly having problems. My favorite issue is when they change a component but don't tell anyone. Nothing like having the guy in charge of our custom build scrambling to figure out what the fuck they changed (try getting the info out of them!) and then update our custom build so we can get 200 machines out that are already a month behind schedule because Gateway forgot to ship them when they said they would or shipped them to the wrong place or the first shipment arrived with the wrong model (always a lesser model!) or arrived with a bunch of broken machines (it has happened). I can't imagine how many hours we have wasted dealing with Gateway's fuckups, for some reason management won't drop them though - probably because we had just as many problems in the past with the other big manufacturers. Of course they won't go with the on-site assembly of whiteboxes, you know why? Because someone with your same misconceptions has authority in the decision!

Re:wishful thinking

[MBGMorden]

"Whitebox" not on par quality-wise versus OEM? I know you must be joking (or trolling). OEMs are popular for one main reason at large corporations: when something breaks you get the convenience of a single point of contact for warranty service. Other than that, they are generally built with whatever lowest cost crap they manufacturer could get their hands on (ESPECIALLY Dell, which in recent years has just gone to crap. It's a pathetic thing when I can pickup a desktop system unit and by lightly twisting on each ends with my hands see the system FLEX back and forth).

You also have to deal with their horrendous OS loads which nearly cripple the system and the fact that 99% of the time they don't include an actual OS install CD to start from scratch - just a "restore disk" that will start off with a fresh system just as messed up as when you started. Heck even when they do include an install disk it's often missing stuff from the full version.

Yes, a whitebox system can be crappy too, because you have the option of buying crap components. However, a custom/whitebox system built with good memory, a reputable mainboard/graphics card/etc, will beat an OEM machine in quality and stability every time. The only caveat is that you have to be preparted to do some legwork regarding the warranties if something breaks (which in my experience, isn't often. only things I've ever had actually break were hard drives, and most of those were from that horrdendous IBM 60GXP line).

Re:wishful thinking

[MBGMorden]

Stability wise, the Athlon64 chipsets are indeed as good as any I've seen. Just a few days ago I had to shut down my Windows XP PC to install an extra hard drive I'd bought (because 550GB obviously wasn't enough, I had to get another 300GB drive :)). My main system (which is actually running Windows XP; I've got another AMD machine running Gentoo Linux) had been up without a reboot for nearly 4 months before I took it down a week ago to add an extra hard drive. Before that I had shut it down to upgrade the cpu and ram. With it sitting on a UPS, the machine pretty much only gets rebooted during the ugprade process :).

Re:wishful thinking

[toddestan]

I have found "whitebox" systems to be no more less reliable than OEM systems. You're kidding yourself if you think that the major OEMs do a whole lot of testing of their configurations. Heck, go and open up a bunch of Dells that are an identical model - chances are you'll find they have different video cards, different brands of harddrives - even different motherboards. I'm pretty sure that Dell just buys whatever is cheap that week, slaps it into a box, and out the door it goes.

With that said though, I have had a fair share of problems with AMD based systems - mostly due to shitty chipsets and crappy motherboards (especially on the Athlon XP line). Since many whiteboxes are AMD, and most OEMs are Intel, that undoubtably leads to the perception that whiteboxes are less reliable. But the Intel PIII/P4 whitebox computers I deal with have been extremely reliable and problem free machines.

Re:wishful thinking

[tomstdenis]

"I believe that a Windows driver fix was made available, which did something silly like disable the +5V line."

Um ... you can't disable the +5V line. That's where things get their power from. Your mouse, usb sticks, keyboard, etc wouldn't work.

My bet [other than you're making this up to troll] is that you bought returned items which were damaged. EVERY SINGLE computer I've bought that was AMD used the nvidia series of chipset. They all worked properly and never had a "plug usb in and die" problem.

I've seen some MSI boards [using VIA chipsets] have random SATA problems and what not. But generally ASUS and Gigabyte are good to go with. My 2P Tyan board is running two high end Opterons and everything on it, even the audio codec, is working perfectly [using nvidia nF4 Pro chipset].

Tom

Re:wishful thinking

[GiMP]

> My bet [other than you're making this up to troll] is that you bought returned
> items which were damaged. EVERY SINGLE computer I've bought that was AMD used the
> nvidia series of chipset. They all worked properly and never had a "plug usb in
> and die" problem.

The chipset was an AMD 768, a first-generation chipset for the original Athlon processors. Checkout page 14 of the following document.
http://www.amd.com/us-en/assets/content_type/white _papers_and_tech_docs/24472.pdf

Potential Effect on System: Data corruption leading to eventual system hangs and failures. Long term component reliability problems due to bus contention.

Suggested Workaround: Disable the internal USB controller. An external PCI-USB adaptor can be used if required.

Resolution Status: Fix planned for future silicon revision.

Re:wishful thinking

[GiMP]

Meanwhile, in my experience, I've rarely had a failure of an OEM machine, while I've had failures in all my whitebox systems to date. For me, it is piece of mind, and also "someone else's problem". With my wife, for instance, if a system I built was to break for any minor reason, it would be my fault -- if a machine from Dell breaks, it is Dell's fault. I don't like being at fault, if I don't have to.

Re:wishful thinking

[tomstdenis]

Athlon != K8/Opteron.

Though my experience with K7 stuff was fine too, only significant problem was overheating.

I seriously think your impressions are just out of date. If you are buying brand name K8 components today and they're not working you're either

a) setting it up wrong

b) buying non-brandname parts

c) buying it at a place you shouldn't trust

d) Lying.

I've bought several 754, 939 and 940 motherboards from ASUS and Tyan and they all work just fine. The fact that you seem to be having consistent "problems" with them suggests that your process requires updating.

Tom

My review

[Anonymous Coward]

This architecture is a pretty big disappointment.

Sure, Socket 939 was amazing when it came out. Nine hundred and thirty-nine pins -- quite an amazing figure.

But I'm sure I wasn't the only one who was less than thrilled at the release of Socket 940. I mean, just one more lousy pin? That's only a tenth of a percent improvement on Socket 939! One wonders what AMD was thinking.

And now here we are, with this farce of an architecture. Yes, Socket AM2 has only 940 pins as well! All that AMD has done in all this time is to shift the pins around on the CPU, much as the occupants of the Titanic spent their time rearranging their deck chairs -- even while that one guy kept on yelling "Iceberg ahead!"

Of course, AMD has tried to hide their laziness with the snazzy marketing name AM2. And yeah, I have got to admit that the name sounds pretty damn good. But in the end, isn't the socket itself more important than a mere name?

Re:My review

[OneManCongaLine]

Of course, AMD has tried to hide their laziness with the snazzy marketing name AM2. And yeah, I have got to admit that the name sounds pretty damn good. But in the end, isn't the socket itself more important than a mere name?

Snazzy name, yeah? I totally misread the headline as A2M...very Freudian indeed!

Re:My review

[Anonymous Coward]

Snazzy name, yeah? I totally misread the headline as A2M...very Freudian indeed!

You need at least cum laude in pornography to be able to freudian slip AM2... Well, I did too. :(

Re:My review

[Anonymous Coward]

Socket 939 came out after socket 940. It was designed to allow AMD to sell cheaper processors without eating into their expensive range of socket 940 processors. It became really really popular, so the expensive FX series of processors became available in socket 939 format. For cheaper.

Word is the socket 939 to socket AM2 switch was designed to make sure people didn't mix up DDR and DDR2 memory with the wrong processor.

Re:My review

[Anonymous Coward]

What makes you think more pins = better?

Have you heard of this technology called SATA?

As far as not being exciting anymore, AMD is working on technology to make multiple processors appear to the workload as one, helping to overcome the non-linear performance gain we currently get by adding more processors to your system.

I think they are keeping their future technology to themselves at this point.

Re:My review

[Bert64]

Socket 940 actually came out before 939

Another benchmark of the AM2

[yorugua]

Here's another "benchmark":

http://enthusiast.hardocp.com/article.html?art=MTA 2NSwxLCxoZW50aHVzaWFzdA==

Socket consideration

[wysiwia]

Introducing a new socket (AM2) and not supporting the existing ones (939) after such a short period doesn't show much competence. I used to upgrade the CPU of any of my computer at least once during the life cycle of 4-6 years but that won't be possible with my 939 mainboard. That means at least one less CPU sell for AMD.

O. Wyss

Re:Socket consideration

[Gothmolly]

Seeing as how both Intel and AMD have changed sockets at least once every 4-6 years, I'd say that your statement is a lie.

Re:Socket consideration

[ErikZ]

Yeah, I can see where spending an extra 60$ to double or triple your PC speed is just unreasonable.

Where's the sarcasm tag button?

Re:Socket consideration

[zaibazu]

If you are so competent then explain us how to design a socket that is efficient for mass production for a CPU with integrated memory controller, capable to connect to different types of memory.

Re:Socket consideration

[Repugnant_Shit]

And I'd say more like 2 or 3 years. I have a socket Athlon 850, a 1.3 Tbird, and an Athlon64 3500+. Each one uses a different socket. I got the 850 in...2000 maybe. So that's a max of 3 years for each? And it isn't like all the 939 parts are going to vanish over night. I'm still thinking about upgrading my 3500+ to a dual core, since my board supports it.

Re:Socket consideration

[Bros]

It is not the socket per se, it's the pinout of the CPU. Considering this, you sure have a point. It is not really practicable to design a CPU with 2000 pins and using only a subset of them, depending on the connected hardware (memory, etc.).

Re:Socket consideration

[GiMP]

Don't forget the new memory. Usually they tend to change the memory when they change to socket as well, and the dual-memory boards are usually from PC Chips or other quality manufacturers, like ECS. Right now, my basement is full of EDO, SDRAM, DDR, DDR2, etc. Some ECC, some not, some motherboards don't work with ECC, some don't (officially) work without ECC. I have some older and more exotic memory as well.

Am I complaining? Not much, I understand it is necessary to improve the architecture. Nn the other hand it isn't fair to say that a processor upgrade is only a processor and motherboard. It is often the case that new RAM is required as well.

Re:Socket consideration

[gdog05]

I bought the ASRock Dual 939 with the AMD2 upgrade slot about 7 months ago. I feel very satisfied with my purchase. I haven't found out how much the upgrade card is, but the mobo itself was only $60+. It's nice to have the feature to easily and cheaply upgrade. Go ASRock.

Re:Socket consideration

[wysiwia]

Before the 939 board I had 2 socket A boards, both had a 900MHz CPU now one has a 2GHz the other a 2.4GHz CPU. Then I had a socket 7 board which I upgraded a K7-II to a K7-III+ (450MHz). I don't remember the early Pentium CPUI's I had upgraded before.

O. Wyss

Re:Socket consideration

[wysiwia]

It's not very complicate to create a CPU which sports different memory systems like DDR versus DDR2, any chipset can do it. Even the current CPU have to deal with different memories so it's quite possible the new CPUs already have this feature. So not supporting 939 is just a marketing and not a technical decision.

O. Wyss

Re:Socket consideration

[zaibazu]

What particular CPUs with integrated memory controller do you have in mind ?

Re:Socket consideration

[tomstdenis]

Wow, where did you get your EE?

DDR2 and DDR1 are NOT electrically compatible. That means you plug DDR2 into a DDR1 processor and it goes boom.

And as for making some form of DDR1+DDR2 processor that's really expensive given the protocols aren't the same, etc. You'd be wasting a lot of die space that will be effectively off all the time.

DDR1+939 chips will be out for a while I imagine. If you're really paranoid you should wait till they get past AM2.

Tom

Re:Socket consideration

[MacroRex]

I bought that board last week. One big selling point was that it has both AGP x8 and PCI-E x16 slots, meaning I don't have to upgrade also my video card before I want. It's a sweet mainboard.

Re:Socket consideration

[scharkalvin]

The existing socket 939 dual core cpu designs will probably
be available for a while yet. So you will be able to
upgrade. Consider that you can't use DDR2 memory in DDR sockets,
so you would STILL need a new MB even if AMD kept the same
socket for the new cpu.

Re:Socket consideration

[wysiwia]

Sure, I'll see in about 2 years.

DDR2 has been known long enough to be considered when AMD designed socket 939. So they are either incompetent or cheaters. Either one annoys me so I probably don't stick with them any longer unless they are factors better.

Why do computer companies never understand to not annoy potential customers.

O. Wyss

Re:Intel's roadmap to the Cornroe and beyond

[Enderandrew]

It really isn't fair to compare the next-gen Intel offerings (just now coming out) against the three year old K8 core which is what people are doing. AMD's next gen processors are rolling out this summer. Let's try to compare apples to apples.

When we compare apples to apples (the P4 line to the K8 line) Intel actually uses more power, generates more heat, runs slower, and is more expensive.

I'm glad Intel stepped up and made a good processor. The Core Duo is a good processor, don't get me wrong. But for all the marketing buzz about Intel right now, AMD does in fact have a response.

AMD is not only rolling out Turion X2 series processors, but they've got the AM2 lineup such as the 5000+ X2, the quad core series, etc.

k8l

[Joe The Dragon]

When amd K8L chips come out intel will fall again also K8L quad cores will be true quad cores not 2 duel cores linked by a fsb also with some K8L chips each core will have it's own L2 with shaded L3.

Re:[offtopic] 'Print' version is split into 10 pag

[Bilestoad]

Even Shorter Summary:

AMD are in deep shit once Conroe appears in volume, as comparisons have already shown that even an overclocked FX-60 loses badly to Conroe. Oh well, it was a nice couple of years for them.

Re:Intel's roadmap to the Cornroe and beyond

[powerlord]

Exactly. The other thing that most people miss is the strengths for each player.

Even if the Core Duo turns out to be better than the comparable AMD chip, Intel still has ground to catch up onin the multi-chip arena due to memory bandwidth. For most people at home this might not be an issue, but for servers it can be.

Between that and the quad-core chip, it would be very interesting to see Intel manage to recapture some desktop space while AMD gets into the Server Room. :)

Re:Moore's law

[somersault]

the parent poster didn't make any claims as to what Moore's law was, he just asked if it was still holding. As his /. ID is only 5 figures long, I'm guessing he actually knows Moore's Law is about transistors and not clock speed. Even I know that. Now stop trolling :p

Better Reviews

[Google85]

There is better and more balanced reviews here and here

Re:[offtopic] 'Print' version is split into 10 pag

[myurr]

Are they really? Conroe has been shown to be quicker in limited tests overseen by Intel. Note that, as far as I am aware, none of the 'independent' testers so far have been responsible for setting up any of the test machines. And while I am sure Intel did not do anything untoward, can you honestly guarantee that absolutely everything is fair and equal in that test?

Take a look at this article: http://www.dailytech.com/article.aspx?newsid=2487

It's showing the new woodcrest chips to be somewhere between 5 - 15% faster, as opposed to Intels claimed 33%.

Finally, have you considered that AMD may just be working on something new. For a start this is their first AM2 chip, they will surely start ramping the core clock as well as the memory clock soon. And they have kept very quiet about their other future developments, not to mention any process changes (ie. catching up with Intels 65nm lead).

So while Intel may indeed recapture the top performance slot for a while it's likely to be transient until AMD release their next big thing. Which will capture the performance crown for a while until Intel release their next big thing. And so on.

Re:[offtopic] 'Print' version is split into 10 pag

[myurr]

Of course I forgot to mention that Conroe is 32bit, the AMD64 chips are... wait for it... 64 bit! You may not feel that is important, but many readers on this site do.

Re:Intel's roadmap to the Cornroe and beyond

[LehiNephi]

I was excited for Conroe, too...until I found out that the Intel-sponsored public benchmark demonstration of Conroe vs. an FX-60 was (surprise, surprise) rigged. More recent, hopefully more trustworthy reviews have still shown Conroe to lead AMD's current processors, but I'll hold onto my money until it actually is released and more reviews are available before I blow them on a new X2 or Core 2 Duo.

BTW, does anyone else see how the name "Core 2 Duo" could lead to confusing people? Joe Sixpack is going to have a hard time telling the difference between a Core (1) Duo and a Core 2 Solo. And is a Core 2 Duo the same as a Core Quad or a Core 4 Solo?

Re:[offtopic] 'Print' version is split into 10 pag

[jiushao]

Of course I forgot to mention that Conroe is 32bit

Well, that's true, except it's not.

Much Better Review

[Enderandrew]

http://www.planetx64.com/index.php?option=com_cont ent&task=view&id=215&Itemid=14

There you go.

What about Pacifica?

[jimcooncat]

Is Pacifica included with this? Can we finally run unmodified Windows under Xen?

Re:What about Pacifica?

[adam.dorsey]

According to the chart at http://www.planetx64.com/index.php?option=com_cont ent&task=view&id=215&Itemid=14&limit=1&limitstart= 1, yes, these processors have virtualization.

More about it later in the review, also.

Re:What about Pacifica?

[Massacrifice]

Yes, all AM2 Athlons (not Semprons) now support Pacifica.

Ah...

[Christopher Rogers]

...Memories of DDR2...

That's what I would do too

[jeffmeden]

...If i had a magical chart making machine that knew the difference between a decline underway or a true bottom. No way would I use it on the stock market or anything like that. Seriously though, you just need to pay attention to product releases like this, and look at how it compares to existing hardware. Big price cuts generally only happen when bigger and better things roll out in the same segment. Price cuts will happen repeatedly, so don't bother waiting for the 'bottom'!

Re:Intel's roadmap to the Cornroe and beyond

[westyx]

See, I like comparisons which compare products that are actually out, because comparing product a with (hypothetical) product b inevitably ends up with "product b rocks! shame it isn't out yet. can't be bought. benchmarks will be x. probably. if everything works out"

It's totally fair. You could say that the upcoming amd chips *might* be better than what intel is rolling out now, but to say it isn't 'fair' smacks of fanboiism.

desktop benchmarks?

[sgt scrub]

I have no idea why they are looking at this in a desktop centric way. In no way is x64 bit or duel core processing support available for important desktop applications(games).

We didn't see any performance improvement by the use of DDR2 memories instead of DDR. In fact, Athlon 64 FX-60 was faster in several situations, showing that at least for the software we used having a bigger L2 memory cache is better than having DDR2 memory instead of regular DDR.

This processor and socket is for servers. I want to know important things like: How long before support is available for Linux or Solaris.

Re:desktop benchmarks?

[pla]

I have no idea why they are looking at this in a desktop centric way.

Because even if AMD didn't prove the clock-rate-only wars as a sham, Intel destroying the P4 with a chip sucking a quarter the power did.

Modern machines run dozens, if not hundreds, of processes simultaneously. Even if not running multithreaded apps, the average user will benefit from going multi-core.


In no way is x64 bit or duel core processing support available for important desktop applications(games).

64 bit (easily) allows more than 4GB memory, end of story. Currently that doesn't matter. Wait until Vista comes out.

Multi-core kicks serious butt entirely on its own merits. If you ONLY play CPU-intensive single-threaded games, for now, a faster single-core will give you more bang for your buck. If you do anything more than just gaming, you'll very quickly see that dual core gives a HUGE speedup for considerably less money than continuing to play the clock-speed war. Now, I suspect going from two to four cores won't make nearly as much difference as going from one to two, until more CPU-intensive apps (including games) start making use of them. But that will happen, and probably within the next two or three years. Now that dual-core has gone from "extremely rare" to "the majority of new machines", developers will use them.

Re:desktop benchmarks?

[sgt scrub]

Games are the only applications that ever promt someone to buy a high end machine. Unless they are planning on developing 3D applications or convert a pr0n shop of movies to mpeg this processor is pointless. As you say "probably within the next two or three years..." software will be ready to use these capabilities. Two to three years in hardware time is 5 to 7 years or 1 to 2 upgrade cycles for gamers.

Re:desktop benchmarks?

[binary paladin]

"Games are the only applications that ever promt someone to buy a high end machine."

What are you? 14-years-old? Perhaps 16? Perhaps you've never had to use your computer as anything but a toy, but for those of us who use machines for work, there are plenty of other applications that prompt upgrades. High end digital photography, print work, 3D modeling and annimation and let's not forget video editing. That's just the tip of the iceberg.

For me, dual cores rock because they greatly assist with virtualization. For someone who runs multiple OSes at once, the speed increase is appreciated. It means I don't need to use multiple computers for testing anymore. One will do the trick of three. This is especially awesome when that one computer is now a small laptop that I can take pretty much anywhere.

Servers, video, and professional applications drive a large portion of high end hardware. Games are only a slice of the pie and they're also only ONE KIND of computing.

Re:desktop benchmarks?

[sgt scrub]

way to read my post.

Games are the only applications that ever promt someone to buy a high end machine. Unless they are planning on developing 3D applications or convert a pr0n shop of movies to mpeg this processor is pointless.

vs.

3D modeling and annimation and let's not forget video editing"

First AM2 roundup

[Marcel_CHW]

Here is a review with four AM2 processor: FX-62, 5000+, 4600+, 4000+

I think is a little more complete than the one in Hardware Secrets, but is in Spanish, any way you can see the graphics.

Re:Moore's law

[Anonymous Coward]

Dang, -2 for a simple question. That's just mean.

What "whitebox" builder are you using?

[Ayanami Rei]

You apparently don't deal with competent people for your 1U/workstation needs.
* Supermicro
* Monarch
* Polywell
* PenguinComputing
* eRacks
* ASA

Take your pick. All of these 'whitebox' vendors release quality, SUPPORTED configurations using custom, bleeding-edge hardware for build-it-yourself prices.

One thing people are neglecting

[Anonymous Coward]

Intels Conroe beats the AMD in speed per watt and speed per clock.

However!

1) it's a whole new generation; AMD's K8 is 3 years old already.
2) Intel uses their new 65nm process, meaning faster and cooler processorts, AMD is still on 90nm for the next 2 months, and then by August they'll be at 65nm too, and beating intel in power per watt again. They've got a 35W server processor on the roadmap; That's incredible.
3) AMD's next generation chip is due out soon too. No doubt it will put them out on top again.

Saying Conroe is an AMD killer is a little disingenious; that's like saying a Pentium 60 is going to beat a 486DX2/66. While it's true, that really isn't a valid comparison.

Re:One thing people are neglecting

[Anonymous Coward]

AMD has lost the performance war for now. Deal with it. It's only a corporation, just like Intel.

Concerning nVidia chipsets:

[Ayanami Rei]

The nForce 2200/2050 MCP is a much better part than the nForce 4/SLI. Its my favorite system baseline for new servers. I love being able to have dedicated PCIe 4x lanes on each HT link and such. In that configuration there aren't many drivers to speak of that you need. forcedeth works fine for networking, it even supports much of the TCP offload capabilities...

An even better mix is the AMD 8131 + nForce 2050. That gives you PCI-X AND PCIe, dedicated. It's my favorite server platform.

Ummmmm.

[GmAz]

Our results, however, aren't final, since we had 1 GB DDR2 memory installed but 2 GB DDR on socket 939 platform. We will redo all the benchmarks as soon as we get two 1 GB DDR2 memory modules and publish here the results.

Why didn't they just test with 1GB of ram accross the board? That would have make it seem fairer than 2GB and 1GB systems.

My two cents is that wait for the AM2 socket to progress further and don't jump on the band wagon just because its there. I am a computer enthusiast and love to have the best hardware in my system. The problem is that with new stuff coming out all the time, there is no point to upgrade just yet. My AMD64 3200+ runs great, does everything I need it to do on my system and more. I would love an AMD64 X2 4000, but the funds are low.

Re:[offtopic] 'Print' version is split into 10 pag

[Anonymous Coward]

when you split the print article into 10 pages, you've crossed line from greediness to stupidity.

Worst webpage layout I've seen since 1998. They should pay for their next review equipment.
It looks like a three guys who got together for a 9th grade project where each asked a question:

#1 - If it's 495 words, is that close enough?
#2 - Can we doublespace it to reach 10 pages?
#3 - Is this for a grade?

Re:[offtopic] 'Print' version is split into 10 pag

[Cornflake917]

Can you link me to the article that says the Conroe is 32bit? Thanks.

The Once and Never Champion

[Catbeller]

AMD has beaten Intel's offerings for, what, three years now, and STILL they can't get a break. Apple won't use them. Dell just this month tentatively offers AMD chips for their server line.

The chips are cheaper and are faster than Intel's. You couldn't tell from the press!

No matter what AMD does, the next line in this type of story is "But Intel's next CPU, expected in the year 2121, is expected to outperform AMD's best. Is AMD doomed? "

Re:The Once and Never Champion

[jkerman]

sorry man, the intel yonahs are running on-par with AMD's latest desktop chips, at ONE THIRD the power. AMD won the price/performace war hands down, the performance/watt war is the next one, and intel currently has the lead.

for how long, who knows?

Re:The Once and Never Champion

[ak3ldama]

65nm has a lot to do with that. amd is 90nm now and 65nm by years end. intel's core duo is 65nm now and that them helped immensely.

Re:The Once and Never Champion

[Burning1]

AMD is taken very seriously by small and home builders. It will take longer for AMD to regain the trust from OEMs and even longer from corporations.

Back in the 90s AMD was the company you went to for budget machines when reliability wasn't an issue. It takes a long time to recover your image from that.

Re:The Once and Never Champion

[chriso11]

Well, let's see if Intel really delivers. Intel has a long history of great promise and failure to deliver. Remember how they were going to dominate graphics? Or how about how wonderful Itanium was going to be? Remember the 1.12GHz P3? I am going to wait and see - if the Conroe really is the next champ, it will be. Intel has to prove itself - the days of Intel simply decreeing each CPU change is over. AMD has delivered what it has promised for the past 5 years. If this makes me a fanboy, well, it's because AMD has earned it in my mind. Intel has to do more than simply announce the next flavor of the month to get my attention.

Re:The Once and Never Champion

[Emetophobe]

The chips are cheaper and are faster than Intel's. You couldn't tell from the press! Faster yes. Cheaper no.

Re:The Once and Never Champion

[Anonymous Coward]

For some definition of "beaten", anyway.

From what I've seen, AMD may be faster if you don't care about any other parameters, like size or heat or cost. In the laptop space, the Pentium M has been more than holding its own, and the Core Duo is taking over for it. The Core Duo is a dual core CPU at under 25W; that's a bit better than AMD's best, the 55W Opteron.

AMD looks very much like a one-trick pony: they do fast workstations and some bigger laptops, and that's about it. Intel does small embedded chips (XScale), laptops, workstations, servers, motherboards and chipsets, graphics chips, and so on.

I've grown up hating Intel, and everybody loves cheering for the underdog, but in this case, isn't it true that Intel is just better?

I mean, if Apple didn't care about laptops, they could have stuck with PowerPC.

Re:Intel's roadmap to the Cornroe and beyond

[Sebastopol]

That's ironic, because prior to Conroe's dominance over AMD's AM2, AMD fanatics liked to ding Intel for being a 10 year old core because they reverted from the Pentium4 core back to the original Pentium III core.

So basically the opposition argument goes like this:

(pre-Conroe)
"AMD's K8 beats Intel's 10 year old core, I guess Intel doesn't have any new designs" ...to...

(post-Conroe)
"Stop comparing AMD's 3 year old K8 core to intel's latest and greatest... um, despite the fact it's still the same 10 year old core..."

I believe that's called a double standard. But again, this is /.

Alway's upgraded mobo + cpu + ram

[Nazmun]

I've always upgraded those three things together. If not at best your getting a 30% improvement in processing speed.

Even if amd used socket a for a long time they migrated from pc100-133 ram to pc 2100-2700 (drr 266 & 333) and bus speeds changed. That required new motherboards. It's quite pointless to upgrade your machine to a new cpu if your bus speed and memory speed remains unchanged in almost all cases. The speed difference is so minor.

Re:Intel's roadmap to the Cornroe and beyond

[ak3ldama]

This argument shouldn't exist. Checkout the numbers that AMD X2 4600 put out, very on par or close to that fx60 and the 5000. That chip is out now and you can put it into your computer you buy from new egg on the cheap. Compare that to what Intel has out now and life aint so peachy. And I am not the type to knock on the intel p3 core, it rocks. It was a good move to go back to it and work forward again. In the end, the AM2 chipset (with 400/800 fsb now, but 3 faster ones coming out later) and the soon coming/core duo intel chips will both be good choices, and present a healthy & competitive market.

it's totally fair...

[YesIAmAScript]

We're comparing the latest from each competitor. That's completely fair. If AMD doesn't want to look bad in comparison, they need to catch up on technology.

From what I understand, AMD will not have 65nm chips until December. That means they won't be catching up soon.

When we compare apples to apples (the P4 line to the K8 line) Intel actually uses more power, generates more heat, runs slower, and is more expensive.

Apples to Apples will change next week when Core 2 Duo comes out. And already, if you compare Core Duo to AMD's chips at the same clock speed, the Core Duo is faster and uses less power.

(comparison against Socket 940)
http://anandtech.com/mb/showdoc.aspx?i=2750

But here's the sick part. Read hothardware's review. They include a Pentium Extreme Edition 3.73GHz in their tests. It runs neck and neck in the tests (except in gaming where it gets shellacked). But the final test is power consumption. At the wall socket at least, the AM2 CPU appears to use more power than the Pentium EE 3.73GHz.

If you can't do better on power consumption than the most power-hungry P4 out there, you should go to the back of the line.

It's clear AMD is behind Intel at the moment. I really hope they catch up and pass them. We've benefited from the competition and I don't want to see it end.

Re:it's totally fair...

[Vlad_the_Inhaler]

At the wall socket at least, the AM2 CPU appears to use more power than the Pentium EE 3.73GHz.

wtf? I want independent confirmation of *that* one. It flies against everything we have seen in the last couple of years. Was there something wrong with the AM2 which meant it ran flat out all the time?

I know they have a problem (design flaw?) with dual-core processors whereby if one processor is running flat out then the other is as well, regardless of load.

Re:it's totally fair...

[Enderandrew]

I read six reviews on the AM2 line yesterday, and they all said the new AM2 processors used considerably less power than the 939 processors, which used considerably less power than the baseline P4 series. So I'm calling bullshit, and quoting an article to prove it. And anyone who argues clockspeed doesn't know processors. You don't compare directly at clockspeed. You compare processors at similiar prices and performance. Clockspeed is the most deceptive thing in the world. AMD 5000+ Max Thermal Power: 130 W http://www.planetx64.com/index.php?option=com_cont ent&task=view&id=215&Itemid=14&limit=1&limitstart= 1 Sorry, but try again. Intel does not currently have a processor on the market that beats AMD.

Re:Intel's roadmap to the Cornroe and beyond

[Anonymous Coward]

People are comparing the latest they can get their hands on, seems natural to me. Do you actually mean that review sites should refrain from comparing Conroe to the latest available AMD CPUs?

"this summer" is next summer. K8L will be released in the middle of 2007.

P4 is older than K8 so that is not a good comparison.

Why do slashdotters have to be in denial mode so often ?

Re:Uneven Benchmark--FX-62 Review

[Anonymous Coward]

Why did they pick a review comparing the 5000 to the FX chip? The FX-62 is the upgrade of the 60, not the 5000. Check out this, more even comparison: http://www.extremetech.com/article2/0,1697,1966062 ,00.asp

Sweet Revenge

[Anonymous Coward]

Good. Enjoy the pain that I felt as an abandoned socket 940 user.

Socket 939 was just an artificial marketing move to segment server and home markets. And don't feed me the baloney about 939 being the only one that supported unbuffered RAM because that's a lie as well (you don't subtract one pin to offer a feature ---that's a function of the improved memory controller on newer chips which socket 940 can also support).

Socket AM2 will meet the same fate when socket F shows up. Artificial market segmentation for no real purpose.

How many errata are fixed?

[Anonymous Coward]

Most important perpetually unanswered question.
How many major K8 errata are fixed?

I learned the hard way I can't trust my mobo maker (ASUS) with BIOS updates since they dropped SK8V support after only 2 non-beta BIOSes. So I hope AMD is finally fixing the ones they said they would. We can't depend on BIOS or kernel workarounds to be a crutch.

Infamous errata #93 for example.
Or chip kill support for servers?
Having major errata on the memory controller is scary. Doubly so on servers.
What about the cache performance feature Linus himself disabled in the 2.6 kernel because it was broken in hardware?

Re:[offtopic] 'Print' version is split into 10 pag

[Anonymous Coward]

Is it possible to write a Firefox extension to create a single page out of multi-page articles like this? There's an extension called NextPlease that navigates to the next page on articles like this, but could it be linked up to something that concatenates them all together?

Why just DDR2 anyways?

[Jesus_666]

Seriously, why are we still stuck with DDR2? I want a processor that supports at least 7th Mix!

how kind.. conroe not even released

[Anonymous Coward]

don't predict demise.... the other product is still vapor until Apple ships it.

Re:how kind.. conroe not even released

[arsenic0]

Apparently you didnt know that Intel released Conroe's to be benchmarked, nice and convienent that they did it on AMD's release day no? :) Good marketting if you ask me..kill all the hype about AMD's release day and make it about Intel... http://www.hexus.net/content/item.php?item=5692 And this is with a gimped Intel board running at 677mhz instead of 800, and only 1Gig of ram not 2 Gigs like the FX62...

Re:[offtopic] 'Print' version is split into 10 pag

[Anonymous Coward]

You are confused about a lot of things, fanboy.

1. Woodcrest is a server part although it's rumored that Apple may use it in desktops. (Woodcrest desktops, Conroe minis and iMacs? Yes please.)

2. Intel's tests at IDF are not the only tests out there. Look harder or wait and see. But I think you will choose to disbelieve anyway.

3. Conroe (Core 2) is 64-bit - http://en.wikipedia.org/wiki/Intel_Core_2 - but why anyone cares I don't know. 4GB memory? Hmm, very few need that in a desktop. Faster? Well, unless there's no driver for some vital bit of hardware, which is certainly the case with A64, then it's useless.

Intel's new products will really create a lot of conflict for many Slashdot readers. On one hand, geek lust - the best available. On the other hand, produced by the 900lb gorilla of the industry - oh noes!

Finally, if AMD had anything to dampen the buzz Intel is getting over Core 2 they would have used it. AM2 is nothing special, just desparate marketing. AMD really are in trouble and they know it, just compare the die shrink schedules - they are going to get less competitive on the desktop, not more. Now multi-core server is another matter, but the longer-term picture there doesn't look so rosy for AMD either.

Re:Intel's roadmap to the Cornroe and beyond

[Anonymous Coward]

In what way was it rigged?
In no way...

It is pretty obvious now that Conroe/Woodcrest are integer monsters, and very good at floating point too. Thus it reflects what Intel said in the beginning of the year.

Re:[offtopic] 'Print' version is split into 10 pag

[Anonymous Coward]

... just desparate marketing.

Well done! That's a clever way of saying that the marketing is both disparate and desperate.

Oh, wait. You just don't know English?

/Damn Foreigner

WRONG LINK

[default luser]

I could have sworn I linked correctly.

Here is the correct link:

Anandtech's Real-Time Pricing Engine

Shocking news: 1/2 as much ram may be slower

[nick_davison]

They found no system speed difference between:

1 GB of DDR2
2 GB of DDR1

If you skip over to their auto department, you'll also find that, despite expectations to the contrary, a Ferrari performs no better than a VW Beetle. Granted, they didn't have any tires on their Ferrari at the time of testing but that's not going to stop them announcing their findings now with a quiet footnote about retesting later.

Re:[offtopic] 'Print' version is split into 10 pag

[ak3ldama]

Wow. I didn't realize what is going on until now. This is just the resurgence of the Intel fanboy. Which is coinciding and colliding furiously with Apple fanboys. Intel has sucked until Pentium M, and now doesn't suck so much with the Intel Core Duo. The price/performance/wattage of machines is becoming equal whether AMD or Intel.

The funny part is that it just hasn't happened yet, but the Intel fans have been waiting for so long that they've just gotta get up and sing anyways. AMD has only a 90nm process, and is still competing with the futuristic Conroe. AMD is still the best option on the desktop. AMD is still the only real option in server land. Want 4 sockets (or even just 2)? Want a reasonable cooling/power bill? AMD is the only option. Intel lucked out by keeping the Pentium 3 architecture around thus managing to keep the laptop/mobile market. Now they're putting those benefits into their desktop/server platforms and by doing being competitive again.

But Intel has not won the battle yet, is the Conroe out yet? Where is the massive proliferation of desktop motherboards for building Core Duo machines? They're not here yet.

Once DDR2 moves beyond 400mhz, AMD will have some serious bandwidth. Later this year AMD will have 65nm, this will bring huge results for their power/watt numbers. So both sides of the story can spout off about what they will have 'eventually.' People need to sit back and re-evaluate things. All we have right now is a healthy, competitive market.

Re:[offtopic] 'Print' version is split into 10 pag

[Anonymous Coward]

No, those 90nm chips are not competing with the 65nm Intel chips. Unless you're you, and you're choosing not to believe all the benchmarks, both Intel-supervised and third-party, that can be found through our favorite search engine. AMD might get to 65nm after Intel has gone to 45nm, or slightly before. They really need to do better than that.

Where are the desktop motherboards? Well if you'd done any research you would know that Core 2 requires a 975-based board, available for some time now, and perhaps a BIOS update, like the one I had to apply when I bought my current A64 3000+ "Venice".

AMD already has serious memory bandwidth. Go to any of those reviews mentioned and the FX62 shows about double Core 2's memory bandwidth - but it's not enough to be competitive this time.

Four sockets is a situation where the memory bandwidth does tell, and I don't know of any Intel response to that - so for now AMD is king in that space. But two sockets means woodcrest, and recent reviews show it as much faster than dual Opterons for lower price and lower TDP. Please, check your facts

Anyone who actually compares price/performance/watt will find your post amusing. Back on the lithium for you, come back to reality.

Glenda Slagg

[Hal_Porter]

Intel? Don'tcha just love 'em? Conroe is wiping the floor with AMD's aging Athlon 64 Chips right now at 2.6Ghz. Intel will really put the MEGAHURTS (Geddit?) on AMD with the 3Ghz Core Extreme. And Woodcrest and Merom should be 20% faster.

Intel? Don'tcha just hate them. Faced with the failure of the Netburst, they've dug out the aging Pentium 3, and revamped it a bit. Now they're releasing benchmarks for unreleased chips, claiming they'll be faster than a high end AMD64. Doesn't anyone remember Intel leaking outrageously exaggerated benchmarks long before a chip release to limit sales of a competitor's faster chip that's actually available now.

Re:[offtopic] 'Print' version is split into 10 pag

[Anonymous Coward]

"We didn't see any performance improvement by the use of DDR2 memories instead of DDR"

And here I was thinking this change was for monetary reasons not performance.

Due to volume DDR2 is now cheaper than DDR, this gap will only increase as more manufaturers switch DDR factories to DDR2 production.

Re:[offtopic] 'Print' version is split into 10 pag

[ak3ldama]

Well, Mr. AC, For the most part you had an intelligent and balanced response. That's nice to see on /. these days. But lets check some stuff. The 5000+ is only using 400mhz ddr2 ram. this will get higher in the future. What is the conroe going to run, 667 or higher i'd assume. from the hexus.net link on /. a little bit ago the conroe beat the athlon quite well at far cry but not so much so at quake4. considering which is out now though it's not really an issue.

btw, did you notice the numbers the amd x2 4600+ put out? that processor is the real one to buy. i would much rather spend $550 on the 4600+ than over $1K on the 5000+ or about as much on an FX62. how much does the conroe cost? the core duo t2600 (2x2.167) is of very similar performance to the 4600+ for $642 on newegg.com.

for power consumption, the core duo t2600 idles at the same wattage as a turion. under load the amd 3800+ was 60 watts away, and the pentium 4 based xe chips hitting 315 watts. these numbers will likely change though once the intel puts out their desktop chips. along with that performance should also increase which is good for intel.

i hope people don't get offended by what i'm saying, i certainly don't want to get put on lithium again. the core duo is definitely worth buying, but not any more so than and amd64. you need to look at the whole solution, not just the cpu. i would definitely have to say that amd has sorta slumbered within the last year or so, and that has allowed intel come back swingin'.

you don't deserve a response...

[YesIAmAScript]

Here's the hothardware link:

http://www.hothardware.com/viewarticle.aspx?articl eid=822&cid=1&page=21

Look at the graphs at the bottom, the increase in power when the CPU is loaded is presumably due to the CPU. The EE adds 80W when going to peak power, the AMDs add almost 90W. Make of it what you will. To me it looks like the AMD is a hog. I say this because I know the EE is a hog.

I dunno why you list 130W as AMDs TDP. Are you considering that a good thing? That's an assload! My entire system doesn't take that, counting inefficiencies. (and I have an Athlon X2).

Clock-for-clock, Intel is beating AMD with Core Duo. In performance per Watt, they are KILLING them. AMD does win out on the top end, because Intel doesn't have high-end chips out yet. This will probably change when Core 2 Duo comes out.

See link:

http://anandtech.com/mb/showdoc.aspx?i=2750

The Intel chip beats the AMD in every test that measures CPU performance. It only falls to even or a bit worse in tests which test primarily the speed of peripherals.

If you are looking at an Athlon X2 4400+ or below and you don't need 64-bit, you're making a big mistake buying an AMD. The Intel is superior. That's why I say "clock-for-clock". If you need a processor in the speed range that Intel covers with Core Duo, the Core Duo is the winner. They just top out at 2.16GHz (almost the same speed as my Athlon X2 4200+), so if you want more performance than that, you have to go to AMD (because going to P4 makes zero sense).

If what we hear about Core 2 Duo is correct, AMD won't have much left to crow about next week. Intel will have 64-bit in Core 2 Duo. AMD willl be beat on performance/Watt. They'll be beat on performance/MHz. They'll be beat on performance in multiprocessing situations. They'll be beat on performance "at any cost" (max performance). If they're lucky, AMD will keep the performance/dollar measure. That's a pretty big fall for a company that had a clear edge just a few months ago.

I don't have anything against AMD. My primary machine is an Athlon 64 X2 4200+ and I love it. And it was exactly the right thing to buy at the time. But that's changed for the mainstream today, and it look like it'll change for the top end too next week.

I wish AMD the best. I hope they have something up their sleeve to top Intel again. But AM2 isn't it. I really hope they can get to 65nm (and presumably competitiveness on power) sooner than the December they state.

here it is again...

[YesIAmAScript]

http://anandtech.com/cpuchipsets/showdoc.aspx?i=27 62&p=11

The FX-62 taking the same power (at the wall) as the P4 EE under load.

Very sad for AMD. They've resorted to performance though power-sucking, the same thing everyone crapped on Intel for with P4.

Re:here it is again...

[Vlad_the_Inhaler]

Ah, 'Under Load'.

The good thing about that (from AMD's POV) is that a powerful 2X desktop processor will not be 'Under Load' very often, the bad thing is that those figures probably also apply when only one of the processors is running flat out.

also another note..

[YesIAmAScript]

Something I didn't notice initially, this is a 65nm version of P4 EE. So it likely takes a lot less power than the usual hog 90nm versions of EE.

Still, I think it does indicate how AMD's architecture is a bit behind the times, and they are stretching it with clock speeds, just like Intel did with P4.

Hopefully (like Intel finally did) AMD will debut a new architecture soon that changes this.