Shot #1 is a standard Quake II engine skybox at the 256 X 256resolution afforded by Voodoo cards. Shot #2 is the same skybox on a TNT based card and utilizes a 512 X 512 image. Notice the difference? This is another example of the numerous scalability options in Soldier of Fortune that allow you to customize the game to run optimally on your system.

A direct comparsion can be found hier.

A word of warning to any unsuspecting gamers looking at purchasing a bargain TNT2 card:

If it's very cheap and not a big-brand name (Creative Labs, Diamond, etc), make sure it isn't a "TNT2 Model 64" configuration. These are starting to flood the stores over here, and are selling like hot-cakes.

There is one problem: they are MUCH slower than the default TNT2 setup. Even within the 40%-50% range in some games. This may be plenty for the average gamer with a low budget. But for those looking for something faster than their old V2 or TNT, they might find it won't gain them much speed at all.

So if the labelling has anything like "TNT2 M64" or "Model 64" mentioned, then be aware you are buying an inferior product to a typical TNT2 card.

Note: TNT2 M64 only owns a 64Bit Bus compared to 128Bit of normal TNT2 cards. More about that here (june news).

"Once again the results, based on Mercury Research's testing, speak for themselves," stated Dan Vivoli, vice president of product marketing at NVIDIA. "Mercury Research results are a key metric of our goal to provide users with the absolute best visual experience on the PC."

In an online report, Mercury Research reported the top performing 3D and 2D processors on the market using Ziff Davis' WinBench 99 test and 3DMark 99 by the Futuremark Corporation. NVIDIA's RIVA TNT2 produced the highest 3D score in both 3D benchmarks and outperformed the nearest competitor by 90 points on 3D WinBench and by 128 points on 3DMark. The testing platform for this data series was an Intel® Pentium® II 500 MHz CPU running on an ASUS motherboard.

Presseerklärung

PG: How will the advent of T-buffering effect the future of PC gaming?

Viet-Tam: The information released by 3dfx suggests to me that T-buffering is really nothing new under the sun. Basically when you want to do things like true full-scene antialiasing, motion blur, depth of field, you need to render several slightly different images and combine, average, or blend (same thing) them into a single image. From the beginning, OpenGL (for example) supported this by providing an "accumulation buffer"; an application can render several images into it, each image adding to the existing contents.

The downside to this is that it takes longer to render a single frame of animation, because you have to render the same geometry several times into the accumulation buffer to get the final displayed frame. What T-buffering seems to do is allow these multiple images to be rendered in parallel (at the same time) into separate buffers (collectively called the "T-buffer"?). Then some hardware reads these buffers and combines them together to produce the displayed image.

"We Mainboard manufacturers are always in a gerneration change. The Nvidia TnT was the first graphic card that was hard on the limit of the AGP specs and which was sensitive about that. Now TnT2 and Voodoo3 are on the absolute limit. But these chips are still inside the specs. How a mainboard manufacturer designs his boards and which parts he uses is his own decision.

If there is a too weak Voltage Regulator design on the board you can´t handle the power consumption of the cards when initializing. Sometimes you can also find inferior parts on some boards. Our new Mainboards P6BX-A and P6Pro-A+ do not have any problems with that. The older P6B40-A4X also does´nt have any problems with these cards. We at Tekram take care that our boards are tested by every branded graphic card manufacturer to do compatibility testing. We do also run our own tests here.

We can only recommend customers to ask the graphic card manufacturer if the card you plan to buy does work with a special mainboard. A good PC dealer should also be able to do this service."

PG: There are alot of gamers still confused about the upcoming T&L(Transform and Lighting) technology said to be incorportated in your upcoming chipset, mainly surrounding performance increase and less cpu consumption. Could you possibly enlighten us some more on why we "need" this technology in the pc games market?

Nick:There's a lot of people who've been talking about how a good xform and lighting engine can off-load the CPU. To give some rough guesstimates, a game like Quake2 or Quake3 probably spends something between 15% and 20% of its time transforming vertices. That's time that the CPU really should be spending doing things like better A.I., or nicer particle systems, model physics, etc. There's no games out there that use lighting now. One of the big reasons is that in order to make lighting look good, you need a lot more geometry in a scene (since lighting is applied at each vertex). But, if you add more vertices, then your CPU needs to get bogged down lighting all those extra vertices. So instead, many 3D games chose to follow the likes of Quake2 and use lightmaps. While they are in some ways great approximations, there's other ways where they don't look good, and cause things to run slower than need be.

PG: Will games have to be written specifically to take advantage of the T&L technology, or will all OGl/D3D games automatically use it as they please?

Nick: The Direct3D API has changed somewhat to accomodate features like hardware transform and lighting, and applications will need to change to take advantage of that. For some apps, the change will be fairly trivial. For others, it may be harder. Well-written OpenGL games would automatically be able to take advantage of faster driver transform pipelines (like Quake2 and Quake3), while there are others (sorry, I won't name names) that perform their transforms in the game code, instead of in OpenGL. Those will need to be modified to take advantage of any hardware that could potentially do transform and lighting. Most of those games are ports to OpenGL from legacy APIs that some 3D graphics companies put out before OpenGL and D3D were as popular as they are today.

Read the whole stuff here.

"Thanks that you give us the chance to comment that. It seems that modern graphic cards, as our Maxi Gamer Xentor32 go to the limit of the AGP-Specs in terms of power consumption because they need high voltages. Mainboards that fully comply with these specs do not have any problem. But there may be problems when the mainboard can´t handle the power consumption. We offer a small sofware (comment: Power Sprinter) that lets the user reduce the power of a Xentor32 (so he can reduce the power consumption) but this is only a improvised solution. Users should prever powerfull mainboards at all."

[prev] [next] [random] [list]