Slashdot Mirror

Except GPU's are at least as high transistor count as CPU's these days and based on transistor count growth GPU's will exceed CPU's in the next one to two generations.

Current GPU's are already ahead of current CPU's:
Amd X2 1mbit cache: 233.2 million transistors
Intel Core 2 Quad: 582million transistors
Nvidia 8800 GTX: 681 million transistors

Do you have some benchmarks of it against the (newer) Intel GMA 950 (which is the chip the MacBook has)?
Why yes I do! Last one is in German, but numbers are still numbers regardless of language.

Given a choice between the two, I'd actually take the Intel chip because it's got a Free 3D-accelerated driver and the Nvidia one doesn't.
That's like saying "No, I'd rather walk than drive. I'd have to pay for gas."
it's got a Free 3D-accelerated driver
Oh, and the driver is always free. It's the chip itself that cost money.

The Q means Quad. This is a dual-core X6800 CPU, not quad-core.

It's an Intel Core 2 Extreme X6800, not an "Intel Core 2 Duo QX6800", which is a non-existent part.

The Falcon NW has a Intel Core 2 Extreme QX6700.

http://anandtech.com/cpuchipsets/showdoc.aspx?i=28 66 has the info.

Its not an either/or situation though. BlandName Ltd. supplies silicon for IBM, they make money. Samsung asks them to supply them with silicon, they do, so they make even more money. Just because IBM gets silicon off them doesn't mean Samsung won't. Where, exactly is the zero-sum part?

Wargames intelligence, which Rumsfeld had access to, predicted the loss. That is why Rumsfeld quit multiple times before today (Bush refused Rumsfeld's resignations on each prior occasion). Now it seems, with the Army Times, Navy Times, Air Force Times and Marine Corps Times all calling for his resignation, Bush has finally conceded defeat. Rumsfeld must go. But his replacement, a person directly connected to the Iran-Contra Affair, is as laughable a choice as Bush having chosen his own personal attorney to be on the United States Supreme Court.

As Rainsford (not Rumsfeld) wrote:

It is EXTREMELY well documented at this point that virtually every expert who looked at the situation concluded we'd need far more troops, and a far better plan, before we invaded Iraq. This myth that "nobody saw it coming" has been well and truly busted, the only people who "didn't see it coming" were Bush and friends, and that was mainly because when the actual military experts tried to talk to them, they inserted their fingers into their ears and started humming.

Going over this stuff, even if there is new evidence, isn't really going to accomplish much. Anybody who doesn't believe Bush and Co were intentionally living in their own little fantasy world when we invaded Iraq is some sort of diehard fanatic that won't change their mind unless Bush himself tells them to.

I really like Google Earth

What the Mac has going for it is Apple's polish on top of FreeBSD's heritage of Unix stability and power

http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2866

The Merom chip is dual-core and has ludicrous amounts of power-controlled cache, which turns off when not in use, and the Centrino platform is highly efficient. AnandTech tested an MSI S271 recently, which is bested for computing power by an Asus notbeook with a 2.0GHz Core 2 Duo chip but, being a 12-inch ultra-portable using integrated graphics, the MSI runs longer than the Asus which has GeForce Go 7700 graphics.

The new MacBook Pro is what I wanted in March when I bought my Turion 64 notebook.

... if you combine them on the same die with a large shared cache and the on-chip memory controller... you can see where I'm going with this. Think of it as a separate CPU, just printed on the same silicon wafer. That means you only need 1 fan to cool it and you can lose a lot of heat producing power management circuitry on the video card.

Obviously this is not going to be ideal for high end gaming rigs; but it will improve the quality of integrated video chipsets on lower end and even mid range PCs.

When Timna was initially finished, RAMBUS was still so expensive that Timna's release had to be delayed so that a (PC100-to-RAMBUS) memory translator could be added. Those of us who followed chipsets back then know how badly RAMBUS and memory translators bombed. The integrated RAMBUS memory controller had to be the biggest reason Timna was cancelled. This might also be a reason Intel doesn't integrate a memory controller onto their current CPUs.

Interestingly, Timna was the first project of Intel's new Israeli design team. Not a great start, but their second project was pretty darned good (Pentium M/Centrino).

quality costs more... spend less time maintaing...

Are you implying that Apple hardware is better? Correct me if I'm wrong, but it certainly seems so.

I'm a native PC user, but I've used macs a lot, particularly in recent years, and I own one myself. The Software is better. No questions asked. The hardware.... well it looks cool. And I have nothing else good to say about it.

Ironically, my 15" Ti powerbook is one of the worst designed computers (laptop or desktop) that I've ever owned. This is quite sad, because I was very impressed with it at first, but it had a massive amount of design flaws (I'll leave the one-button-ness out of this; I usually use a USB mouse anyway). I bought it in 2003, when about 99% of laptops sold had USB 2.0 (the other 1% or so being Apples), but no big loss. The first minor problem happened after about 3 months, when the latch to my Powerbook occasionally popped open. I didn't see this as a major problem at first, but it did start getting more frequent. It didn't perse affect the usability of the laptop, but a lot of times it would pop open and drain battery life, and one time it poppeed open and started running (very hot mind you) while in the case.
 
Just after a year passed (conveniently right after the warranty was up), my Superdrive jammed up and wouldn't accept any optical discs. Shame. There was nothing "super" about it in the first place. It was basically a DVD burner that costs $150 instead of $60 because it says Apple on it. Happens quite frequently so I hear, but Apple doesn't seem to care if it's past warranty.

Despite the sturdy Titanium shell, the frame on the side of the laptop is flimsy plastic, which ended up cracking in several places. Did I drop it? No. Just from picking it up occasionally with one hand (gasp). The weight of the laptop was enough to bend the plastic enough to fracture over time. Also, the cord started coming apart from the piece of plastic that I where the AC adapter plugs into the back of the laptop. I am not the only one that has had this problem (read the reviews).

As for price/performance, Apple has a very large history of being WAY behind the curb. When I got my powerbook (1 Ghz was the fastest), most P4 laptops at the time were around 2.0-2.2 Ghz. More efficient chip blah blah, there is no wsy, by any stretch of the imagnation that any 32-bit 1.0 ghz will beat any 32-bit 2.0 processor. In fact, the PowerPC isn't even that much faster even at a 1:1 ratio clock speed. And on top of being slow, still costs more? I am not particularly impressed.

I have to admit, Apple has made a LOT of progress moving over to the Intel platform; they have gotten a LOT faster (with considerably reduced power consumption too). Of course, now I fail to see how the mac hardware is fundamentally different from standard PC hardware. The architecture is the same now, and based on my experiences (and others'), it is certainly not better (Don't forget the passively cooled cube "supercomputer" either). I think Apple should suck it up and open up their platform a bit more, instead of suing the asses off of anyone that tries to do it for them. I'm certainly not saying they should phase out their hardware, but they (and their fanboys) really should stop putting it on such a pedestal.

This post is turning out to be a bit more of a troll than I intended, and I'm not certainly not anti-Apple, but I really feel like they should be making a few adjustments. In the future, I'd really like to see Apple focusing more on software (OS X is unquestionably the best OS out there), iPods, and services like iTunes.

they are absolutely some of the nicest machines I've ever seen

You mean the machines that overheat and pollute the environment? The machines that run a fundamentally flawed operating system? The power of rationalization accorded the distracted-by-shiny-plastic crowd never ceases to amaze me.

So since we are talking games...

The top of the line iMac comes with a 7300GT Nvidia card. You can up it to a mid range 7600GT card at most. Now with Windows installed, do you think that the 7600GT would drive the 24" monitor as well as oh say a dual 7900GT SLI setup?
Oh, get a Mac Pro you say. Well except that you still have no SLI or Crossfire support and extremely limited choices of video cards. And it costs an arm and a leg. Sure it's cheap for a dual Xeon workstation - but if you just want to play games you are better off with a slower CPU and a top of the line graphics card setup.
Oh and the FB-DIMMs in the Mac Pro severly impact gaming performance. Check out the gaming benchmarks and you'll see that the Mac Pro comes in last in every one. The 3GHz Xeon (both dual and quad) gets beat out by a Core 2 Duo running at 2.66GHz.
Mac hardware is absolutely NOT the superior choice when it comes to high end game playing.
I use and like Mac OSX but one thing I dislike about the platform is the lack of hardware choice. That's the reason why you can build a PC for gaming that will be higher performing than a Mac and will cost less.

Nonsense. ......... slide film is better than print.

Hello? Say what?

Just for your edification though, it is generally accepted that given current technology, the difference in dynamic range is still about 1.5 to 3 stops better with film than digital. Clicky clicky for just one reference out of many.

If you have a look around, the X1900XT actually uses about 110 watts peak, for most board suppliers. The X1900XTX only uses about 120 watts.
There's a few places that confirm this, here they show just the VGA card usage, here they show the total system power usage.

A few weeks ago Anandtech already tried to plug two 2.4 GHz Quad-Core Clovertons (Xeons) samples into the new Mac Pro featuring two LGA-771 sockets. Worked like a charm, a nice eight core machine. And since dual socket motherboards are quite expensive, the Mac Pro might even be a cheap version.

5) Upgrade the CPU. A real bitch

how does my case look? it's got 3 windows. 3!!! of course, the window on the side with the motherboard tray is pretty useless...

and, it may not be the smartest decision, since they get hella scratched up, but i've always been fond of the clear design

The 7600GT clobbers the 6800GS in every benchmark. Hardly the "30% perfomance increase" you were talking about, unless you're referring to the 7600GT's advantage over the 6800GS. . .

See http://www.anandtech.com/showdoc.aspx?i=2717&p=10

Gigabyte's 4GB ram drive... and apparently 8GB wouldn't be technically hard for them to make. Though, a number of their tests show a negligible performance increase.

They said Gigabyte told them that the drives were unstable in a Raid 0 for w/e reason... Maybe they have it fixed. If they do, and they come out w/ an 8gb version, that'd be a cool 16GB.

It is talking about operations/sec. A better (and more valid-sounding) analysis is on Anandtech, which says this is basically an engineer's toy and not something they're pushing to really develop:

http://anandtech.com/cpuchipsets/showdoc.aspx?i=28 40&p=6

So they don't really want to make this into the P5, they're just playing around with reduced-instruction processors to see what happens.

Um, OS X already can. They just don't sell the boxes yet.

http://www.anandtech.com/mac/showdoc.aspx?i=2832&p =6

not only that, but i decided it was time to get a dual-core because i've started doing multiple things now that i have a pc that's capable. in the past, converting video or extracting, compressing, etc meant leaving my machine to do it's business (or just surf the net). now, i can encode a dvd image for easy access on my file server and play a game. considering i only have the weekends (while in school) to really use my tower, i want to get the most out of it.

personally, i don't expect to be upgrading for a while, but that all depends on what sort of new toys amd/ati comes out with, i'm curious to see what sort of motherboard/cpu/gpu/something solutions show up in the market in 2007.

i don't maintain any loyalty to any one company, i go wherever i see the best deal. when i considered the price/performance as well as the power consumption (add up the extra electricity used in a month), i saw the core 2 as the winner by a great margin. although, i am slightly chagrinned to admit that the gpu i chose is the 1900xt (for the electricity), but i remember seeing a few reviews that pitted it against some of the other top'o the line gpu's in various video situations (dvd's, encoding), it mopped up.

I hope no one uses this as information to buy their motherboard for the new Intel Core 2 Duo if they want a tandem graphics solution. This "chipset" comparison is completely devoid of real-life issues. Like what happens if you want to build a tandem-graphics card rig. The e6700 CPU is a very desirable part right now and if you want SLI... Oops none of these chipsets support SLI... Ok, Im not partial, Ill buy an ATI solution. Only there isnt a real good one. Of the two fastest crossfire cards out now, one is slower than snot the other is a dual-slot! So if you want a good crossfire solution, youll have no PCI-X slots left... So the solution? The Asus P5N32-SLI SE Deluxe Motherboard. Why? It supports all Core 2 Duo parts at full speed. And even the memory controller handles the 1066Mhz speed better than the 965/975X motherboards! Doh! (See http://www.anandtech.com/ Yes, the P5N32 SE may not handle the new quad core parts, but I dont know it wont either. Tom's Hardware just built a $10,000 dollar rig and posted it on their website. Which motherboard did they use. Gee, I wonder... Ad a sidenote, all AMD has to do is get there 65nm process going, Vista will be out.. and Intel will have, by comparison, a very obviously poor solution, bucause there CPU's wont be any faster in XP and AMD's 64 bit capabilites will outshine Intel's when it comes to Vista.

I thought that there must be some problem with the system if they're unable to get all the CPUs under full load.

It's actually really easy to do if your memory system isn't meant to service 8 cores. And the article pretty much backs this up, every time the quad cores fail to shine it's blamed on the memory.

Also, when the "Quad" system failed to shine, it was not blamed solely on the memory. The only benchmarks where the "Quad" didn't shine were the ones that didn't take advantage of the extra two cores, so the extra bandwidth of FB-DIMMs were also not being used.

We saw this same situation when dual processor (single-core CPUs) systems first appeared. At the time, many benchmarks were single-threaded and even some operating systems didn't take advantage of multiple CPUs.

So, I think regular DDR2 @ 667 = 5.4 GB/s... divided amongst 8 cores is just 677 MB/s per core.

I wasn't browsing low enough and I missed an AC post before mine giving a good link (http://www.xbitlabs.com/articles/cpu/display/amd- energy-efficient.html). There was also one after my post (http://www.anandtech.com/cpuchipsets/showdoc.aspx ?i=2795&p=7).

Thanks for both of these. I've looked them over and will read them in detail later. The anandtech article covers power only on page 7 (the one linked) I think, and I was pretty disappointed to see the ratio of idle power to full power. More than 60% it seems. That doesn't bode well for my plans to have an HTPC that can stay on all the time. I'd like it to use 5% of the power when it isn't doing anything.

The xbitlabs story is longer and I couldn't find exact numbers I was looking for during a lunch break.

It appears labs compare processors by running a given program and comparing watts (which works for some things like decoding an HDTV stream), but I'd also like to see numbers from a given task and energy measurements (e.g. compressing a video) over the period of time needed. I'll try looking for that too.

Dara

I can always count on Anandtech to demonstrate how useful Adblock is.

Here are what I found useful recently at Anandtech:
*.dealtime.com/*
*.doubleclick.*
*.linksynergy.com/*
*.mediaplex.*
*.smarttargetting.com/*
*.zedo.com/*
http://anandtech.com/fuseads/*

anandtech's power consumption comparison:
http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2795&p=7

In every case, the Intel processor more expensive than the AMD to which they compare it.

It's also worth noting that the cost of Core 2 Duos has been dropping (as Intel ramps up production to meet demand) while the cost of the AMD chips is largerely fixed or rising.

Other price tracking sites seem to confirm this: Intel vs AMD.

As AMD is _already_ severely beaten in price/performnce, this does not bode well for them.

In every case, the Intel processor more expensive than the AMD to which they compare it.

It's also worth noting that the cost of Core 2 Duos has been dropping (as Intel ramps up production to meet demand) while the cost of the AMD chips is largerely fixed or rising.

Other price tracking sites seem to confirm this: Intel vs AMD.

As AMD is _already_ severely beaten in price/performnce, this does not bode well for them.

I don't agree. Just compare E6300 to Amd X2 4200+ for example. They are priced similarly but Intel leads in most benchmark - not just in this test but all the tests I have seen. In many cases E6300 even beats X2 4600+. And sometimes even 5000+. Look at this for example: http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2802&p=4 (They include both stock and OC results)

Put This web site on one of those babies. It'll be maxed out for the next few hours :)

The last Intel processor I bought for home use was a P2. I recently purchased a Core 2 Duo 6400 and I am *so happy* with it. As discussed at this article:

http://anandtech.com/cpuchipsets/showdoc.aspx?i=28 02

The Core 2 Duos are tremendously and easily overclockable. I upped my performance 25% by changing the FSB from 266 to 333. While this sounds like a significant overclock, for the Core 2 Duo it is actually rather conservative. You juse switch to DDR-667 memory. I'm using the stock Intel cooler and my chips are running just fine temperature wise. People who are more ambitious are going for 400+. When you combine the inherent performance and value in the line with the ease of significant overclocking, AMD isn't even in the same ball game anymore.

I am sure they are happy also. Your crappy, pre-standard N card likely screws up their g network. Good neighbor :(

http://www.anandtech.com/showdoc.aspx?i=2824&p=9

Finkployd

until 802.11n routers can play nicely with other wireless networks and not interfere with 802.11b/g WLANS...

The conclusion to Anandtech's review on "draft 802.11n" routers showed just how bad these products can interfere with existing 802.11b/g networks. It's pretty freakin' bad (bold emphasis mine):

In our preliminary mixed mode testing we experienced the "bad neighbor" effect several times. Not only with our own internal 802.11g network but also visits from actual neighbors who were upset with having to constantly reboot their systems during our testing phase. As we stated earlier, the current 802.11n Draft 1.0 products utilize channel bonding to combine two 20MHz channels into a one wide 40MHz channel. Without proper fall-back techniques, this type of channel bonding can basically take over the entire 2.4GHz band that these products utilize. While the current 802.11n draft states that routers should not interfere with other networks in the area there are not any specifics as to how this will occur. At this time it is left up to the individual manufacturers to determine a "good neighbor" policy.

Network file I/O on OS X sucks. Here's why: http://www.anandtech.com/mac/showdoc.aspx?i=2520&p =8

"The split funnel suspect
The last suspect is the locking system. In Panther, only two threads could lock into the kernel to execute code of the kernel. One thread could lock into the networking part, while the other into the rest of the kernel services.

In Tiger, the locking is finer. Although Apple's documents indicate that it is still rather coarse grained, it is clear that more than two locks into the kernel can exist at the same time. In the case of MySQL, this should be a very important improvement, but we didn't see any improvement at all when performing the tests on both Panther and Tiger. This is speculation, but based on our data, we are tempted to hypothesize that the new locking system isn't really working right now, and that Tiger continues to behave like Panther."

I used a DP Powermac with both Panther and Tiger and it's performance was pathetic with network file I/O almost certainly due to this. Any mac home theater solution that requires a network file server is not only unnecessarily complicated, but a poor performing choice. A better choice for a mini would be an external FW hard drive, but what Apple really needs is a pizza-box headless mac for the living room. The mini is too small.

On a similar theme the FF/Moz/Seamonkey tab name for http://www.anandtech.com/ is rather damn funny as well ;)

Actually it does support both DDR and DDR2. Here's an earlier article comparing memory performance between the two. Almost no difference.

When I noticed this in the factsheet today, we quickly dropped our 7900GS in to our Mac Pro to test it out. Unfortunately, it didn't work. We're not sure if it's a driver thing or a ROM thing since we don't really have the time right now to break out a ROM dumper and check things out, but at least out of the box you're going to be SOL.

The review at Anandtech includes benchmarks against cards going back to the 6600/6800 era.

Why is the submitter only pimping the HotHardware review? Here's more (in no particular order):

HardOCP
Guru3D
Anandtech
Bjorn3D
PCPerspective
nV News

Not true. Check out this review of the ASRock 775Dual-VSTA. The PCI Express Slot is limited to 4X, and it underperforms other PCI-E motherboards by as much as 10%, but it's usually closer. AGP performance is actually slightly better. It's one of the most affordable Core 2 Duo motherboards out there, and it can even use your old DDR400 RAM.

Buying pre-built is always costly, which is a lot of the reason why people like me end up going the Windows OS route. You can't by a modern Mac for under $1000

I'd be willing to bet that I could build a quad-core machine out of AMD Opterons that would smoke your Mac.

If you were used to Windows and not OSX then you'd likely not be making claims like you did.

Get a MacMini. $500...

If you read TFA, de-interlacing and scaling are crucial to providing good DVD quality. NVIDIA and ATI provide this with PureVideo and Avivo technologies, which use their recent GPUs, drivers, and updated MPEG2 decoders to give DVD quality that surpasses high-end set top DVD players. The Mac mini provides none of this.

It's not about performance. Read that again. In today's market, it's about the platform and the pricing as much as it is about the product. Core 2 is too expensive ($240 for a 2.13GHz Core 2). Most of the market doesn't care whether or not Intel's $250 CPU beats AMD's $250 CPU (as it turns out, the competition is remarkably close). It's not about Athlon X2 vs Core 2. It's about Sempron vs. Celeron.

It's nowhere near close, first of all. It's just not. Look here and here for proof. (I hate linking THG, but the numbers at Anandtech agree with what they've posted.) And the Core 2 Duo is too expensive at retail, but not for OEMs like Dell or HP. They're the ones sucking up the supply. Considering how aggressively the chips are priced (even the E6300 beats all but the highest-clocked X2s), this is truly dangerous for AMD. You're right to posit that it's the budget market that determines who really "wins," but it's no longer Sempron vs. Celeron, it's Sempron vs. P4, because those prices were slashed to hell. Intel has better performing parts in every price range. That's a very scary thing for AMD to face.

One could say the same thing about Intel. From the release of Athlon 64 to the release of Core 2, nothing that Intel has released for desktop computers has even been close to the AMD equivolent.

But they more than made up for it in the laptop segment--which is where Intel knew to throw its weight, because now it's come full circle. AMD, on the other hand, has made no significant chipset or CPU updates since the X2. The manufacturing capacity is something I did not consider and you're right to call me out on it, but even so, I can't help but wonder what AMD's R&D division has been doing this entire time. It's been eons since the A64 first came out. It's seen plenty of revisions, sure, but when are we going to see a new chip? They need something to even the performance/price ratio.

Weeee... another troll.... Who modded this up?

It's still more Open Source than Windows.

On the server?
On the destkop?
Care to elaborate?
Links perhaps?

Really?!?! Based on all the facts you provided I suppose we will have to believe you!

Weeee... another troll.... Who modded this up?

It's still more Open Source than Windows.

On the server?
On the destkop?
Care to elaborate?
Links perhaps?

Really?!?! Based on all the facts you provided I suppose we will have to believe you!

"MacOS X has infamously bad threading, which makes it an absolute dog for many important server apps. Anandtech, what I regard as one of the most trustworthy hardware sites on the Internet, has an article outlining the problems:
http://www.anandtech.com/mac/showdoc.aspx?i=2436&p =1 [anandtech.com]"

But that can change by the end of the decade, can't it?

"Slower than Linux or Windows? I'd like to see those numbers, please!"

MacOS X has infamously bad threading, which makes it an absolute dog for many important server apps. Anandtech, what I regard as one of the most trustworthy hardware sites on the Internet, has an article outlining the problems:
http://www.anandtech.com/mac/showdoc.aspx?i=2436&p =1

Unless MacOS X for Intel has gotten miraculous improvements in this area, and I'm not aware it has, you'd be an absolute fool to use MacOS X for any server apps requiring high performance threading.

-Erwos

After reading Anandtech and Tom's hardware for years, just looking at the site hurt my eyes.

The crazy thing is that, despite being on a seemingly ancient 90nm process, the AMD chips still seem to consume less power at idle that the core duo, at least according to Anand. That's why I'm currently sticking with my AMD's in my 24x7 boxes, which spend 99% of their time idling away.