AMD Purchases ATI

[PMS Wedge]

What will this mean? Not really sure... but it's rather significant!

http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~110899,00.html

[Look Its a Panda]

Lets just hope that we won't end up with having to choose between Intel/Nvidia and AMD/ATI. I'm not really that faithful to any of the manufacturers, but I've got AMD and ATI in most of the machines here at home, but mostly because they've been best value when I bought/upgraded my machines.

If AMD will make solutions that will make ATI perform better on their systems, this will be a loss for consumers.

[BlacKwidoW!]

I agree panda, I currently have a AMD but a Nvidia card in my machine right now, it will be interesting to see what path this all takes and how it effects things in the future.

[The Geeky Panda]

its good but right now more and more games are being built on teh Nvidia graphics card which is really really annoying us ATI users since we are the ones who get graphic bugs in the game because of it, liek BF2 for example, thats why my new gaming rig will have SLI 7800 GTX

[Rip]

I'm not surprised you brought this up wedgewu, even with the rumors flying around everywhere since may, its still a bit hard to swallow.

So what does it mean for the major players in the markets?
Intel: Of the four, Intel will probably be hurting the most, but its not deadly pain. Conroe and ATI's crossfire setup was a pretty golden pair, and what had been used in early tests from the get-go as SLI wasn't supported from Intel's chipset. Sadly crossfire probably won't be going very far, according to a short blurb by the INQ N A SHOCK TO no one, a few days ago, ATI had its chipset bus licence pulled, or at least not renewed by Intel. I guess that means that AMD won't be making many chipsets for Intel in a year or so
Sadly intel will have to bite the bullet and get chummy with Nvidia, and the two aren't exactly on buddy-buddy terms. But business is business, and with such large market share, you can bet that ATI's cards will be working on Intel platforms just fine for a while still.

Nividia - they recently announced they're happy about the deal. Surprised? Well, its not a huge shock, after all this puts Nvidia in a position where they're the only major GPU contender that won't have ulterior motives infused into their design process. It also gives them a lot of leverage with whom they partner with in the next years. And Nvidia's stock price reacted accordingly.

AMD - This is the hardest one to make a prediction on. Their stock took a tumble today on the news, but in reality if there is shaky investor confidence, the people who would have voted 'no' to the deal have sold out. Regardless of stock price, a bunch of analysts believe that AMD needs this in order to survive. The fact of the matter is when it comes to PC production, selling just OEM processors and having such a narrow focus is a real risk. With this acquisition, AMD can start expanding their offerings with such things as their own chipsets and integrated graphics. In addition ATI has a lot of experience in other things AMD could really use in the long run. Some people are hyping the possibility of a low-end CPU having a low-end GPU integrated into the same package (essentially what PS2 did later in its lifecycle to cut costs). This could provide a cheap solution by AMD for the future. There's a bunch of possibilities, and I think it will benefit AMD in the long run, but it might be a bit tight with a 2.5 billion debt for a while.

ATI - Considering the stock's ~19% jump today, I think people are viewing this as good news for ATI. While AMD said that production will still be done at their fab partners, ATI should benefit from the resources and expertise AMD can offer. I only hope they retain their Canadian spunk, because after all ATI does have a very interesting corporate personality, and I kinda like their attitude recently compared with Nvidia's somewhat overbearing 'you will listen because we have SLI' attitude. That and I happen to like the unified shader concept, but thats an entirely different story.


On a final note, GN Flipinomad - both AMD and ATI work with developers to promote different games. For example, PREY worked with ATI while BF2 worked with AMD. TWIMTBP (The way its meant to be played) from Nvidia has just been hammered into the public eye more. While ATI's drivers have been complained about often enough in the past, Nvidia's latest incarnation is really....well....pretty crappy from what i've seen, and their vista drivers aren't any better. They've taken a page from the catalyst book and took out a lot of the options I have become used to having. Bug crashes are always relative. I've used BF2 on my desktop and my laptop, and while its slow as heck on my desktop due to the spec differences, the graphics card (an ATI 9600XT) doesn't fail at all either. I've played UT2k4, a heavily Nvidia sponsored game, on my desktop back when it was my main machine without any trouble at all. Same with playing prey on my laptop with a go6800. There are some slight performance differences between companies with respect to certain games, but its not nearly as dire as you're making out.

[H2O Silence]

/agree with Rip.

This is quite significant, but not in the way with some people are thinking.

I have seen some interesting things that AMD and ATi have up their sleeve with this, and unfortunately at this time I'm not at liberty to talk about it.

Expect some announcements as to core group projects and future endeavors in the coming months. Although... ATi has a quad-core chipset in working development... Intel surely doesn't have any quad-cores coming out soon. =x

But what do I know, hey? ;)

[asphix]

I'm undecided on how I feel about this so far. Its going to re-define a lot of how the industry works.. and there are a lot of good possible results as well as a lot of possible negatives.

However, right now... I feel more positives will come of this than negatives.. especially considering how Nvidia has been gaining a firm upper hand as of late with their SLI and also considering Intels recent moves. We'll most certainly see more products, lower prices and more innovation from all parties involved as competition heats up.

This also works for me as I'm an Intel guy.. and I love my Nvidia... so hopefully now they'll have to get along and support SLI on 975x.

All we can do is wait and see.. AMD and ATI both have their roots in the enthusiast market, and both are on the forefront of innovation (4 x 4.... Unified shader pipeline... and thats not to say that Intel/Nvidia dont innovate :) ) so it will be interesting to see what they come up with once the dust settles. I see a lot of potential for proprietary GCPU's for use in conjunction with the 4 x 4 if a use ever deems itself worthy of something along those lines.

[PMS Wedge]

From what I understand, SLI is now being supported on the Intel chipsets. :) Good news, for sure! And with those 7950s... quad SLI! (even though nothing takes advantage of it yet)

I WOULD like to understand how 4x4 is "innovation" when people complained that the Pentium D was just two cores slapped together... 4x4 doesn't seem much better. :P But I haven't been reading much on it lately, so please enlighten me if I'm incorrect.

H20 Shaun... were you being sarcastic? Quad-core procs coming from Intel? I'm pretty sure it's been stated publicly that Q1 2007 is when Kentsfield (the quad core Conroe) is launching... which actually will work in the same mobo as Conroe does.

I don't think I'm at liberty to comment on the deal too much more since you all know where I work (though my comments are my own and not on behalf of Intel, I will take precaution in this). However, I will continue to say that it is quite an interesting turn of events. :)

[asphix]

Awesome! SLI on 975x! That would completely make my day.. as Crossfire isnt attractive to me at all. Althought SLI isnt all that attractive either considering the top of the line single card solution is always as strong as the past-gens top SLI config.. and running SLI with the top of the line current gen offering isnt exactly financially friendly.

First off, I just want to make clear that I wasnt one of the people criticizing the Pentium D or even Kentsfields design. 2 cores on one CPU (without shared cache) is a logical step and I believe Intel's plan (http://www.tgdaily.com/2006/07/10/intel_32_core_processor/)of doubling up cores, then re-designing a unified cache, doubling up, redesign.. is a very intelligent business model and will help saturate the market with processors containing seadily increasing core counts @ feasable prices. For both intel and its consumers. Its a very aggressive strategy that simply makes sense, which I personally believe will work out extremely well.

As for the innovative aspects to 4 x 4.. that would mostly be due to Hyper Transport and the possibilities it enables. Rather than type it out I'll copy and paste from this (http://techreport.com/onearticle.x/10091) article.

More exciting down the road are the possibilities created for multi-socket 4x4 systems by AMD's move to license coherent HyperTransport to partners, another of today's revelations. The non-coherent version of HyperTransport is already in use around the industry as a chip-to-chip interconnect technology, but coherent HyperTransport is AMD's own secret sauce, the means by which multiple Opterons in a system communicate. Opening up coherent HyperTransport to third parties could eventually allow for all sorts of coprocessors to plug into the second socket on a 4x4 system and not only talk to the processor in the first socket at very low latency and high speeds, but also participate directly in the CPU's cache and memory subsystem.

AMD mentioned media and physics processors as candidates for being adapted to coherent HyperTransport. These coprocessors could plug into a CPU socket or a HyperTransport slot, known as an HTX slot, to accelerate games and other apps. The tight coupling of CPU and coprocessor could lead to mind-blowing performance in the right types of application. Of course, hanging a coprocessor off of a coherent HyperTransport link will have to make sense in order to be worth the effort. For instance, GPUs already have their own very high bandwidth local memory subsystems and might do just as well on PCI Express. Sound cards as we now know them, meanwhile, probably couldn't take sufficient advantage of better communication with the CPU and memory.


Its more than just a dual processor platform. Theres a lot of potential to expand the market into areas unknown.. and taking that into consideration, the AMD/ATI deal becomes a bit more attractive.

Of course, it all hinges on how well 4x4 does at first, and if people prove there is a market for this. With Kentsfield coming out at the same time now.. that is going to hamper any demand for 4x4... especially since I have a feeling a single Kentsfield will yeild substantially higher performance levels than a comparitively priced 4x4 configuration at much higher efficiency levels (aka less heat disspation.. better overclockability...) not to mention a single chip solution is more upgrade friendly.

4x4's potential for innovation also depend on the development of hardware which would take advantage of the platform. Therefore, its essential that the adoption rate of the 4x4 platform is strong (afterall its not an industry standard) in order to provide incentive for companies to invest in R&D for such a thing. And even with a strong adoption rate, it would take a while to convince developers and consumers that the platform is more than just a gimmic.

So it goes without saying this is essentially why I believe Kentsfield's release was moved forward.. and its also what I believe to be the core benefit of the AMD/ATI deal. AMD gets ATi to develop a hardware solution for a 4x4 system.. no PCIe graphics card.. but a specail ATi chip that shares communication with system memory (lowering overall cost of a system built on this platform) and having low latency communication directly with the CPU and most importantly defining its practicality and negating any thoughts of "gimmic!!!" (lead through example). ATi gets to be at the forefront of this new platform.. which if it takes off could be a very significant position. You have to admit, a platform such as this would be fairly attractive to system builders like Dell and HP, not to mention it finally mark the beginning of upgradeable GPU's.. no more buying brand new video cards with expensive GDDR2 or GDDR3 memory.. etc etc etc. With this, theres potential to have lots of choice in regards to how you build your system, and to system builders platform flexibility is all important.. second only to stability.. thus one reason why I feel Intel and Nvidia have been their first choice for a very long while.

This is all simply speculation on my part (lets call it an educated guess :) ).. so dont take any of it as fact.. but I think you can see where I'm coming from when I say theres a lot of room for innovation at this point. Either way, its a gamble as it would take years to mature into something that would generate significant revenue.. and have an installed base to provide solid excuse to develop hardware for it as the enthusiast segment is just a small fraction of the market. But, you have to plant seeds if you want to have a garden.. and AMD did it with 64 bit processors, as useless as that feature really is to the average consumer.. and has been since its debut.. maybe they can do it with 4x4.

If I had to chose.. I'd choose intels solution.. as I'm not much of a gambler. But I cant deny giving credit to AMD for doing something different and taking a risk.

p.s. sorry for the long posts.. I try and keep them short.. but this is what happens. :( also sorry if some of it is a ramble.. I still havnt completely organized my thoughts on the whole thing.

[asphix]

Thought I would share.. this is an interesting interview with Jon Carvill of ATi. Though he seems to avoid answering any of the questions directly... damn PR people :mad:


Link (http://www.driverheaven.net/dhinterviews/joncarvill/)

[Rip]

I was reading a few interviews today from different reps of all companies. Nvidia does indeed seem to be taking this as good news (same with their stockholders), and for Nvidia fans the fact that Intel is going to be good for quad-SLI ready is pretty big news.
As for the quad core comment, its really a relative term, almost like if you consider quad-sli truely quad sli or just double dual cards...if that makes sense. The definition of a true quad core is debatable, but I for one consider kentsfield to be a legit quad core solution even if the cache isn't shared between all 4 cores (among other things). Oh and its been bumped to late 2006 for kentsfield from all the news ;) I'm excited. I really would like to see what a kentsfield does compared to a 4x4 where hypertransport could play a crucial role in inter-chip comm. I suspect that should help the FX-62 regain some ground on core2....although that doesn't say much in terms of price where I doubt AMD can afford to compete.

Asphix, I don't think we're really going to be seeing much of a change for the gamers when it comes to AMD + ATI. They'll still have dedicated GPUs with their GDDR3 or 4 for a while at least. The advantage of CPU + GPU on one component is really for the lower end, it could provide some pretty nice options for motherboard manufacturers and sales companies such as Dell. Speaking of dell, do you think this move from AMD might have anything to do with a pretty lucrative possibility in the lower end, high volume sales to dell / HP companies?

Intel definitely has had to reshuffle some priorities, but its really not much more than a sting. It won't take their market share down anytime soon.
I'm interesting in getting your opinions on:
http://theinquirer.net/default.aspx?article=33240

Its nice to have some serious business / hardware discussions on here. I really missed that.

[PMS Wedge]

It is quite nice to have some serious hardware discussions. Hence my idea for setting up a board... though we wouldn't generally have it at this level of discussion. I just always thought it'd be nice to have a place where people can bring questions about computers, or problems that they're running into. Or, you can often learn a lot just by reading what other people have to say (there's only handful of us here...)

Anyway, I don't really understand the mentality of something not being "real" dual core. (Or quad core?) Maybe there's some unwritten rule saying that things need to share a cache to be dual core? I don't know. I mean, don't get me wrong - I understand all the tech behind it, but the criticism seems to be a semantics thing. It's two cores on one die/package. End of story.

Asphix, I see what you're saying, but I will leave the term innovation unused until I actually see it in action. On paper it doesn't actually seem that impressive to me, but I tend to be super skeptical about these things (oh so hardware jaded I am..).

Incredibly interesting times in the hardware world now.. :) Definitely exciting to follow!!

[Rip]

Everything is marketing, induce a fake paranoia in the public and you can run away with it (duct tape and terrorism anyone?). Maybe in the tech thing we can get a subforum eventually, continued discussions like this would warent its viability.

There's a good chance 4x4 is more of a paper launch anyway, its just too expensive at this point that very few enthusiasts would want it, and i doubt AMD have a huge stock of 4x4 ready motherboards. I could be wrong though.

[asphix]

I think the main problem (in regards to the bashing of different ways of implementing dual/quad core) is simply due to peoples genral love of complaining and fanboyism. Thats why I ususally disregard comments that in their entirety resemble either. Chances are anyone who complains about seperate cache in Kentsfield will be purchasing one provided benchmarks are good.. the Pentium D line was a good target for the naysayers because consumers were unsatisfied with the performance.. and performance issues had absolutely nothing to do with how the dual core solution was implemented so much as the netburst architecture as a whole.. but thats a discussion for another time ;)

I know my ideas on 4x4 are radical.. but I love hypothesizing and speculating. Its a remote possibility and probably wont be as sucessful as any of my posting made it seem. There is a lot of potential there.. sadly, that potential depends on numerous variables... which pretty much guarentees a strong likelyhood of failure. If Intel wasnt pouring on so much heat, I would say there was a great chance of it suceeding.. but now that core 2 duo is out, and intel has launched its new strategy of providing ever increasing core counts to consumers at an aggressive pace... luck and circumstance will become increasingly important variables to AMD's success.

Personally, after thinking on it some more.. the place where I see this platform really being of use (if it were implelmented correctly and all my hopes of a GPU for the secondary socket became reality) is the mobile market. If there were a way for 4x4 to migrate to a mobile platform in the future, it would finally give mobile users the option of upgradeable GFX. Of course, AMD doesnt exactly have a strong offering int he moble market.. so thats more wishful thinking at this time (along with all of this.. i must admit).

Some more details (http://techreport.com/onearticle.x/10428) on 4x4 have popped up (http://www.theinquirer.net/default.aspx?article=33268)today. You will need to marry the CPU stats and AMD will be providing multiple bundles. They dont look to be consumer friendly though with two of the bundles hitting "under the 1,000 dollar mark" (with the rest.. however many other bundles they will have... hitting above the 1k mark).

Thats too expensive IMO as I expect the entry level kentsfield to be going for around $400.

p.s. I agree.. i love tech discussions.. RIP, in regards to your posted link.. sounds interesting, but I know next to nothing in regards to that company and their products/history, so theres really not much I can provide to a discussion on the topic. Care to elaborate a bit on its significance?

[asphix]

Bit-Tech.net has an excellent article that outlines a lot of my feelings on the AMD-ATi buy-out in a much more eloquent and organized manner. Its very much worth the read, so anyone who wants to have a better understanding of the implications involved, you really should read it.

The article can be found here (http://www.bit-tech.net/hardware/2006/07/24/how_it_all_shakes_out_for_4/1.html) and below are some quotes that stood out.

GPU on CPU: The biggest aspect of this deal is the integration of GPU functionality onto the CPU. Let's examine this.

No co-processor has ever really survived. Remember maths co-processors on the 386? Integrated in the main processor. Remember motherboard cache? Integrated. Memory controllers? Integrated. Floating point units? Integrated. The CPU, it has been said, is basically a black hole that swallows other technology.

This isn't a theory; this is actually happening. AMD has even stated this in conversation with analysts and press this morning. Expect to see integrated GPU-CPUs in 2008.

(Note - the exact form of the integration is still up for grabs. The graphics capability could be a general purpose core on the CPU. It could be a GPU core on the CPU. It could be a dual processor motherboard where one of the processors can be a drop-in graphics chip. That is still up for grabs, since not even the engineers know the best answer to that yet).

and one I found really interesting:

GPU on CPU: Intel has a strategy of integrating GPU functions onto the core of future CPUs. This has been the case for a while now, and shows no signs of changing. We mentioned that there will still be room for uber-high-end discrete boards, but Intel will be expanding its integrated graphics line like crazy. It will be integrated onto the CPU rather than the motherboard, and the capabilities will get better.

Intel's major sales of desktop CPUs have been to business, and businesses now need 3D graphics processing with the advent of Windows Vista, and going forward. Rather than making them buy discrete boards, thus benefitting NVIDIA and ATI, you can bet that business buyers would be happy with uprated Intel graphics directly on the CPU - it would save them money, give them the functionality they need and also keep things simple.

Intel's GPU on CPU is also modular, meaning you can effectively create 'SLI cores' - for example, a 6-core processor could have four CPU cores and two graphics cores working in 'SLI'. This technology could also be spun off into a discrete card if Intel decided it wanted to move into that space to compete with ATI-AMD.

These quotes are only in regards to my above comments.. theres a lot of great, insightful information concerning all the parties involved.. Intel, AMD, ATi and Nvidia. Good stuff.. gotta love bit-tech.

Be sure to check out their forum thread (click "discuss" at the end of the article). Their community is hands down, one of the best tech related communities on the web. Tons of great comments.. you could waste hours reading through it and learning/opening your mind to different possibilities and/or ideas.

[Rip]

Asphix, I wasn't too familiar with it either, exactly the reason I posted it.

While we're on the rumor mill here, I saw another interesting thing on the INQ that is worthwhile for sparking some thoughts. While I never would expect it to happen, apparantly there is another rumor coming from someone saying it might be wise for Intel to do a hostile bit for ATI at $23 a share. link (http://theinquirer.net/default.aspx?article=33265). I'm pretty much of the conclusion that unless something major comes up from the shareholders for AMD or ATI the merger will go through without too much trouble, and I really don't see what exactly intel has to gain going for ATI (although some might say they've got a fair bit to loose ignoring it). But hey, while we're chucking wild ideas on the fire ;)

One of the biggest things that has cropped up recently, at least more so than I remember in the past, is the whole HyperTransport technology AMD brought about, and at the moment anyway, would allow them to do an 8x4 (still named 4x4 of course) without having scaling issues conroe might have with bus saturation due to FSB speeds and whatnot. While its clear Intel has the performance (and at the upper end at least, price / performance) to eaisly top AMD, but as we're scaling up to quad core chips and multi-socket solutions, does AMD have an advantage with their architecture design that can overcome their current performance defecit? And I guess the real question, what will Intel be doing to respond, because since wedgewu still has a job they clearly aren't sitting on their hands ;) (haha, i've got no idea what area of the company you're in, but it made a funny analogy)

Oh and from what ATI is saying, they'll still be doing chipsets for Intel as long as customers want them and until Intel says enough.....which they claim has not happened yet (despite a few rumors to the contrary). I think its good for intel to have some ATI chipsets, although maybe since intel can do crossfire on their own its not as huge of a deal. (Of course there's a good chance they won't want to do crossfire anymore after said merger...but thats a somewhat separate discussion)


Oh also, as to your quoted commentary asphix, I must say I'm in agreement for the most part. Vista is bringing a whole new realm of things to the graphics industry, as this interview (http://blogs.mercurynews.com/aei/2006/07/nvidia_ceo_an_e.html#more) Dean Takahashi did with Nvidia's head describes somewhat....but be ready for some blatent PS3 advertising in it :-x (yes the same Dean Takahashi who wrote Xbox 360 Uncloaked....really good book btw). But essentially, it means that especially initially there will be huge demand for graphics technology upon the release of vista. After all, people want to see the eye candy right? This presents some interesting opportunities and almost a guarantee for a busy year (or two) for nvidia, but also in my opinion sheds some light onto why AMD NEEDS to have more chipset experience in-house on the lower end. If ATI can give them an inexpensive solution far above what VIA can handle, and they still have Nvidia out there, they'll be sitting really pretty for vista, while quite frankly, Intel's integrated solutions at the moment won't pass vista muster unless they really get craking.

[PMS Wedge]

Just as a quick note:

I do performance analysis at Intel, with an emphasis on gaming performance. So I spend a lot of time building rigs and testing them, especially on games. I work a lot with both AMD and Intel machines (since we need to see where our competition stands) so I am quite familiar with both sides.

Now I must run to save the world! (Core 2 launch.. yay!) I'll post some responses when I get some free time. I do really love this discussion though!!

[Rip]

Just as a quick note:

I do performance analysis at Intel, with an emphasis on gaming performance. So I spend a lot of time building rigs and testing them, especially on games. I work a lot with both AMD and Intel machines (since we need to see where our competition stands) so I am quite familiar with both sides.

Now I must run to save the world! (Core 2 launch.. yay!) I'll post some responses when I get some free time. I do really love this discussion though!!

Ok you just made us all really, really jealous! Hey, if you're in the market for interns next summer :p (Actually, intel has a nice booth at the carreer fair every year, stopped by last fall just to find out a bit more about it, have a very nice internship this year with Lutron electronics, but while dimmers are exciting....actually some of the stuff is really cool, i'm working with their commercial systems...complex stuff...but not something I could see myself going into down the road.)
Anywho, I digress.

I must say I'm intrigued as to how the cross-liscencing issues will be worked out between the 4 main companies involved. AMD has some cross liscences with intel and vice versa (instruction sets and whatnot), ATI of course has stuff with both Intel and their new masters, and Nvidia i'm not sure exactly where they sit, but they of course are getting some hefty insider info from both. I'm assuming VIA and the smaller chipset / graphics makers have some sort of deals which are now a bit tangled somehow.

I probably am a bit out of the company life and death cycle, but what happened to matrox? I've got a card from them in my dual PIII 1ghz machine (well server now) and I heard them still up until the last year or so. Are they anywhere in the market at all?

[asphix]

Rip, most your questions (and some of what you said in your prior post concerning the rumor of Intel outbidding AMD on the purchase of ATI.. or even Intel purchasing Nvidia) is touched upon in the full article at bit tech. Read it if you havnt :p you'll be happy you did.

Really, intel doesnt have much to worry about. Its Nvidia whos in a pickle in this situation. Its already known Intel plans to move to an integrated memory controller in the near future.. there just simply isnt a need for it currently (as is demonstrated by Conroe). Once they do that, they will have a scaleable interface similar to Hyper Transport.

The difference that I see, is while AMD spent R&D on an integrated memory controller and Hyper Transport technology.. Intel was focusing on die shrinks (among other things). While both are solid strategies.. intels has positioned themselves at a CLEAR advantage.. as FSB saturation in a design with an off processor memory controller are doing fine for now.

Quad core will be pushing the bus saturation limit.. and whenwe see a unified cache version of quad core sometime Q3-Q4 2007 (or whenever it finally hits.. intel seems to be making a good habbit of pushing forward dates lately) I've a feeling we'll hear announcements of an 8 core, independant cache (per 4 cores) processor on 45nm process sporting an integrated MCH. Intel needs to do this to adhere to their longterm plans in the server segment of competing with the SUN Sparc workstation (which is where AMD got the integrated MCH idea and Hyper Transport to begin with...... they just moved it into the x86 space).

As for the graphics boon with Vista.. I see Intel cashing in more than Nvidia or ATI as intel currently holds 90% of the graphics market (or some absurd number like that) through the use of their integrated graphics solutions. Matrox had a nice try with their.. I forget what the product was called... started with a "P" and it had built in tri-display capabilities (back in 2002 or so.. was sort of a big deal). Sadly, the 3d performance was horrible.. and when you buy an add on video card you do so for one of two reasons. 1) gaming/3dwork... 2) TV/video input/video editing. Matrox is around.. but they are a very niche product mostly for people who need powerful 2D solutions that offer multiple display functionality at a bargain price.

Wedgewu, have fun with the Core 2 launch!

[PMS Wedge]

I got mentioned in a blog!! http://www.dvorak.org/blog/?p=6336

But misquoted - I don't think I could take down the FD no problem... I think I told the guy that it depended on the game. :P Silly journalists...

[Rip]

we just won't talk about what happened when I played FD in some of the xbox games at MLG Philly last year :-p
(oddly enough, havn't played any of em since then...but I digress)