DFI's LANParty UT NF4 Ultra-D motherboard
AMD 3800+
1,5 GB RAM
NVIDIA 7600GT 256MB
250GB WD hard drive
19" Compaq CRT

LG E500 15"
Inter Core2 Duo 1.86GHz
4G RAM
NVIDIA something 256MB
300GB HD

My first laptop actually. Love it

Core 2 Quad Core @ 2.4 GHz
4GB RAM
950GB (2X 150 SATA + 700 SATA)
Nvidia 8800GTS 640MB
Asus P5NE (I didn't buy the SLI version, I think I will be getting a Radeon 4780 X2 though now)
1 LG DVD Burners
Creative X-Fi XtremeMusic
800 watt PSU
ThermalTake Case

+

24'' Samsung Syncmaster 2493HM LCD.

With that setup I can run Crysis decently on high, Crysis Warhead is even better. Most games I run at 1920 by 1200 without a hitch.

Well, It's a little outdated now, (I built it on March 07) but I have:

Core 2 duo @ 2.4 GHz
3GB RAM
800GB (300 IDE + 500 SATA)
Nvidia 8800GTS 640MB
Asus P5NSLI (So I can just drop another 8800 in there when I need to and a better CPU)
two dvd burners
650 watt PSU
Shitsta Ultimate
19" Westinghouse LCD

The only game I can't run on full is Crysis. But I can run it at near-max with about 35 fps.
Anything else I get 70fps. ( I run all games at 1440x900)

So it's decent, and gets the job done well on just about anything. And it's sort of future proof as well.

I haven't watched all of it and probably won't tonight, but I question their logic.

That companies are too zealous in allegations of anti-trust I have no doubt, but that does not undermine the value of the entire system. Anti-competitive price fixing results in a wasteful allocation of resources, and higher prices for consumers. Firms, particularly those with monopoly power charge at the marginal cost equal to marginal revenue price creating a deadweight loss, underproviding, and charging a higher price to reap greater profits. Or at least, that's the basic theory behind it, introducing trade makes it messier.

Not that some industries are not natural monopolies due to falling average costs over the supply of an entire market by one firm, but for the most part, new start-ups should be encouraged. An increased number of firms increases competition, breeds innovation, greater product choice and differentiation to cater to any individual consumer's needs and inevitably lowers prices through knowledge spillage as an externality, even if certain firms are able to hold patents for several years.

Again going by theory, if you're a monopolist for a product that has no close substitutes and particularly if that good is a necessity, then you as a firm excercise significant market power. Realistically though, I would argue a monopoly industry is more a theoretical framework (like perfect competition) that rarely if ever applies to a good in the real world, particuarly when you factor in international trade. Most necessities either fit into competitive industries or like utility companies tend to be natural monopolies.

I don't doubt though, that anti-competitive behaviour occurs. Producers may not be able to hold consumers as dependants on their products (well excluding tabacco and pharamaceutical anti-depressants among certain other exceptions that is ) but they can certainly push out would-be new entrants, by deliberately undercutting new entrants through selling below average cost prices and exploiting their inability to secure long term funding while running consecutive years of losses, or otherwise developing insurmountable barriers to entry. It's not like a monopoly has to continuously keep prices competitive to stave off entrants. After all, exit/entry costs in the real world are substantial in most industries and should a monopoly undercut a would be entrant, the new entrant can't simply get an 100% refund on their investment and hitherto accumulated capital. All that is necessary, is the mere threat that should a new firm enter the market, the monopolist will lower prices to unbearable levels that will stave off any would be entrants. How would you suppose competition is to be maintained without anti-trust laws preventing this from ocurring?

Do consumers demand inovation from monopoly producers supplying consumer wants? Sure. While NVIDIA was dominating the graphics card industry with it's series 8 GeForce cards and AMD/ATI was in a rut, they brought out the 8800 Ultra as a superior high-end card to the 8800 GTX. Since then AMD has gotten back into the game and looks what's happened. Numerous releases of new cards and revisions on previous editions, not to mention the price level of low/mid/upper range cards has essentially halved.

It's a good video, but it is misleading. This was at an nVidia sponsored event, correct? Generally upgrading your GPU is a good help if you're a gamer (and if you're watching this you probably are) but this does sort of suggest that GPUs are just far faster than CPUs and it's not black and white. Technically it's right in that GPUs will do this quicker, but only because these tasks are very simple and there's a large number of them. If you were trying to model a physics theory then a CPU will generally be better at taking that one single task, chewing on it and spitting out one single result as the 196 processors on my new nVidia GTX 260 would spend all their time waiting for some other processor to finish a calculation which it needs the result for. Graphics is different, I game with about 2 million pixels needing processed at least 40 times a second so each processor can largely work on it's own batch of pixels.

The right coder with the right compiler can use GPUs for some funky stuff and CPUs will continue to increase the number of cores on each chip too. I've heard of some research places using game consoles as both their CPU and their GPU are dirty cheap (as MS or Sony is footing the bill, expecting to make it up from commission on game sales).

Anyway the "cpu" had a speed control on it. For all we know they only ever put it up to 3. Crank it to 11 baby!

If it's from an Nvidia conference then I guess I am a bit of an ass because it's somewhat computer related and I suppose actually supposed to demonstrate the power of GPUs. I was just being snarky though, not trying to debate soundly.

In reply to this comment by Aemaeth:
I get the similarity, I just think the presentation sucks. Trust me: I remember when the first 3d accelerators hit the market, I know the difference between GPU and CPU rendering. The reason I say it seems like propaganda is because first of all, you are the first person to use the term rendering. The video title or captions in the video doesn't say GPU Vs. CPU video rendering, it just says, "GPU vs. CPU". Second, realize this is from NVision: nVidia's annual conference. Third, because of the way it is presented it makes it seem like they are trying to say that graphics processors are just plain faster than central processors.

I get the similarity, I just think the presentation sucks. Trust me: I remember when the first 3d accelerators hit the market, I know the difference between GPU and CPU rendering. The reason I say it seems like propaganda is because first of all, you are the first person to use the term rendering. The video title or captions in the video doesn't say GPU Vs. CPU video rendering, it just says, "GPU vs. CPU". Second, realize this is from NVision: nVidia's annual conference. Third, because of the way it is presented it makes it seem like they are trying to say that graphics processors are just plain faster than central processors.

I appreciate criticism for my comments, but not when I get it in the form of an unfounded statement that really says nothing objective at all. Saying, "you must be joking" is really not an argument at all.

In reply to this comment by budzos:
Sure. It's dumb because of two things. First, a demonstration that GPU rendering is faster than CPU rendering is not "propaganda". It's a fact that GPU rendering is much faster. Do you also say that videos are "banned" when they are pulled from the air by the company who created them?

Second, the poster is making a joke. The presentation is not related to computers at all. If you know what "GPU rendering" means, which you must if you think this is "propaganda", I don't see how you can miss that painting with one gun is like rendering with the CPU, and painting with a 1000 nozzles at once is like rendering with the GPU. See now I feel dumb just for explaining it.

In reply to this comment by Aemaeth:
I'd love to hear the explanation for this comment.

In reply to this comment by budzos:
Cool project, but seems like propaganda to me. Why is that a GPU vs. a CPU?

Please be serious. It'd be awesome if you were that dumb.

I have a Nvidia 8800 GTS, I downloaded and installed the CUDA packed drivers to enable PhysX.

I decided to skip all the demo wankery and go straight to testing PhysX in Ghost Recon 2 and Unreal Tournament 3. There was a severe hit on performance with PhysX enabled, I had to drop from 1920 to 1280 to get anything really playable at 60 FPS or so. The effects added seemed gimmicky and didn't add anything to the game atmospherically.

So far not worth it. Though perhaps in the long term, with developer uptake, this will give Nvidia a certain exploitable edge over ATI.

In reply to this comment by marinara:
i bet yer machine would play the second half-life...that was fun.

BUT, after you upgrade get F.E.A.R..... it's an awesome game.... maybe even Quake IV

Actually was able to run those OK. Got HL2 Orange Box, Quake IV, played FEAR. The only thing my machine doesn't really fun are the latest games that came out in the last year or so, ie. CoD4, STALKER, UT3, etc.

It's a P4 3.2 (no duo/quad proc) with 2GB PC3200 and a BFG Tech Nvidia 7800 (still AGP). So it's definitely an older rig, but still handles everything say....pre-'07 pretty good. Eventually I'll upgrade but it won't be for awhile as I'm putting myself through school on my own. I make too much to get grants, etc, but not enough to go full time and get it done and over with fast. I'm thinking I'll have to upgrade for when Fallout 3, GTA4 and Left4Dead hit. Definitely upgrade for L4D.

lol nerds all look alike the world over.
imagine: you could play WoW, AoC and WAR beta simultaneously!

>faster rendering in programs like 3D studio max?
hmmm, i betcha it could. nvidia or ati should be working with these guys to do just that.

Check out NVIDIA's CUDA language that lets developers program the GPU. They give examples of many other tasks that have been done, but basically some tasks lend themselves to parallel execution more than others. It's a good thing these guys in Antwerp didn't assume it couldn't be done since it wasn't being done!

*hypersalivate over that stack of new 9 series geforce cards .. I'm about ready to order another 9600GT, the next time my 7900GTX ruins another round of BF2.

The 9 series feels really good in the hand, well built metal enclosures. Notice they are flush with each other in the video. Do you think Nvidia designed them so their air channels are cooperative when multiple cards are aligned?

This is far from new tech.
Nvidia tried this about 10 years ago on their 3x series cards.
Worked on RS: Rogue Spear. Made your eyes hurt like a mofo afterwards though.

"Minimum system requirements:

Intel Skulltrail chipset with 2x Core2Extreme Quad-Core QX9775 3.2Ghz or better
4 Gigabytes PC3-10600 1333MHz DDR3 SDRAM
3x nVidia 9800 GTX in Tri-SLI config (because nVidia pays us enough to ignore ATI)"

and you'll only get 25 frames per second too!

THANKS CRYTEK!