NVIDIA’s GeForce FX 5800 Ultra GPU

ARE YOU ready? The era of cinematic graphics is upon us.

Err, well, it will be soon.

Or, uhm, maybe not really, but we can show you how it might have looked.

You see, we have our hot little hands on NVIDIA’s NV30 chip, better known as the GeForce FX 5800 Ultra, that was intended to usher in a new era of movie-quality graphics when it debuted some time last year. Instead, it’s months into 2003, and GeForce FX cards still aren’t widely available.

Let’s just say NVIDIA’s had a wee problem with its vaunted execution. That fact hurts even more because NVIDIA’s rival, ATI, hasn’t. The Radeon 9700 debuted last fall to deservedly positive reviews. The 9700 lived up to the high expectations we’d set for this new generation of DirectX 9-class graphics chips, delivering amazing new shiny objects and lots of rippling glowy things onscreen. Err, delivering true floating-point color formats and amazing amounts of real-time graphics processing power.

But the story of the GeForce FX 5800 Ultra isn’t just the story of a late semiconductor. No, it’s also the story of an impressively “overclocked” piece of silicon with a massive, Dustbuster-like appendage strapped to its side. Most intriguingly, it’s the story of a product that may never reach store shelves in any kind of volume, because all signs point to its premature demise.

That said, the GeForce FX 5800 Ultra is still interesting as heck. I’ve managed to finagle one for review, so buckle up, and we’ll see what this puppy can do.

NVIDIA’s tech demo heralds the “Dawn” of cinematic graphics

What’s all this talk about cinematic rendering?
Let’s start right off with the graphics egghead stuff, so we can fit the GeForce FX into the proper context.

The GeForce FX 5800 Ultra card is powered by the new NVIDIA GPU, widely known by its code name, NV30, during its development. The NV30 is NVIDIA’s first attempt at the generation of graphics chips capable of a whole range of features anticipated by the specifications for Microsoft’s DirectX 9 software layer.

The single most important advance in this new generation of graphics chips is a rich, new range of datatypes available to represent graphics data—especially pixels and textures. This capability is the crux of NVIDIA’s marketing push for “cinematic graphics.” By adding the ability to represent graphics data with more precision, chips like NV30 and ATI’s R300 series can render images in real time (or something close to it) nearly as compelling as those produced by movie studios. Now, the GeForce FX may not be ready to replace banks and banks of high-powered computers running in professional render farms just yet, but you might be surprised at how close it comes.

I’ve written more about this new generation of graphics chips and what it means for the world right here. Go read up if you want to understand how such things are possible.

Along with richer datatypes, this new generation of GPUs offers more general programmability, which makes them much more powerful computational devices. Contemporary GPUs have two primary computational units, vertex shaders and pixel shaders. Vertex shaders handle manipulation and lighting (shading) of sets of coordinates in 3D space. Vertex shader programs can govern movements of models and objects, creating realistic bouncing and flexing motions as a CG dinosaur tromps across a scene. Pixel shaders, on the other hand, apply lighting and shading effects algorithms to sets of pixels. Pixel shader programs can produce sophisticated per-pixel lighting effects and generate mind-bending effects like reflections, refractions, and bump mapping. DX9 pixel and vertex shaders incorporate more CPU-like provisions for program execution, including new vector and scalar instructions and basic flow control for subroutines.

Of course, DirectX 9-class pixel shaders include support for floating-point datatypes, as well.

By offering more flexibility and power to manipulate data, vertex and pixel shaders amplify the graphics power of a GPU. Some fairly recent theoretical insights in graphics have shown us that traditional, less-programmable graphics chips could render just about any scene given rich datatypes and enough rendering passes. This realization has led to the development of high-level shading languages like Microsoft’s HLSL and NVIDIA’s Cg. These shading languages can compile complex graphics operations into multiple rendering passes using Direct X or OpenGL instructions. Programmable hardware shaders can produce equivalent results in fewer rendering passes, bringing the horizon for real-time cinematic rendering closer.

So that is, in a nutshell, where all this fancy-pants talk about cinematic rendering comes from. Built on a 0.13-micron manufacturing process with over 125 million transistors, NVIDIA’s GeForce FX is a powerful graphics rendering engine capable of doings things of which previous graphics chips could only dream. The leap from GeForce4 to GeForce FX is as large a leap forward as we’ve seen since 3D graphics has been driven by custom chips.

The big, green, shiny fly in the ointment, however, is ATI’s Radeon R300 series of chips, which brought similar graphics capabilities to the mass market last September. In the following pages, we’ll review the capabilities of NVIDIA’s GeForce FX with an eye toward its competition. We’ll break it down into key areas like pixel-pushing power, vertex and pixel shaders, texture and edge antialiasing, and performance in real-world games to see how the GFFX stacks up.

Cinematic graphics: Lubed up and a bit chilly

ATI counters the nymph with a chimp. Canny.


The card
Before we dig into the GPU performance stuff, though, we should stop and talk about the GeForce FX 5800 Ultra card itself. Words can’t do it justice, so let’s just skip to the pictures:

BFG Tech’s Asylum GeForce FX 5800 Ultra

The passive cooler on the back side of the card is no slouch, either

The Radeon 9800 is smaller in every dimension than the GeForce FX 5800 Ultra It’s big. Double-stacked big. Like an indulgent Oreo for the obese, it takes up twice as much room as a standard AGP card, encroaching on the nearest PCI slot to get room for its cooler.

You may not be able to tell this from the picture, but the GeForce FX card is also heavy. The card’s cooler has more copper in it than a roll of pennies, and using my highly scientific “hold one in each hand” method, I’ve determined the GeForce FX card weighs about the same as three Radeon 9700 cards.

The most infamous feature of the GFFX 5800 Ultra card, though, has to be its Dustbuster cooler. (I coined the term in reference to the FX cooler in this article, thanks very much.) NVIDIA’s “FX Flow” cooler design is reminiscent of the cooler on Abit’s OTES card, which we reviewed a while back. A heat pipe design pulls heat from the surface of the GPU and into the copper fins. The FX Flow’s blower pulls air into the PC case through the lower set of vents, then pushes the air out over the copper fins you see in the upper chamber of its plastic case.

Incidentally, this card and cooling design is not the work of BFG Tech, even though we’re looking at their card. All GeForce FX 5800 Ultra cards are essentially the same, because for this product, NVIDIA generally is supplying its partners with complete cards rather than chips. Likewise, the cooler you see on our BFG Tech card is NVIDIA’s reference cooler with an Asylum sticker slapped on the side. Other NVIDIA partners have cooked up alternative cooler designs for the FX 5800 Ultra, but most cards will probably be similar to this one.

For general non-3D applications, the FX Flow blower on our BFG Tech card doesn’t run at all, leaving the cooling job to the card’s massive copper heat sink apparatus. The copper heatsinks on the card get too hot to touch, even when the PC is sitting in power-save mode, but the fan stays silent, and all is peaceful. Once we kick off a 3D game or application, the blower spins up to speed, emitting a high-pitched whine that elicits flashbacks to my days on the flight deck. And I was never even in the service. After the 3D application ends, the blower will generally spin right back down to a stop. Sometimes it stays on for a while afterwards to bring the chip’s temperature into check.

To give you an idea how loud it is, I used a digital sound level meter (Extech model 407727) to measure the sound of our test system with the GeForce FX and with a Radeon 9800 Pro. The meter was mounted on a tripod about two feet away from our test system, whose only other real noisemakers are a bog-standard AMD Athlon XP retail cooler and a Codegen PSU with a fairly quiet fan inside. (The hard drive is a Maxtor DiamondMax D740X.) I tested noise levels at the Windows desktop and running Unreal 2003. Here are the results:

This thing is loud. Decibels come on a logarithmic scale, so the numeric difference you see here may not capture the difference in noise levels adequately.

Let me be brutally honest here. I hate this cooler. It’s louder than a bright plaid leisure suit, and the white noise repeatedly lulled me to sleep as I was trying complete this review. I didn’t like Abit’s original OTES card, and I don’t like NVIDIA’s expensive knock-off of it any better. Perhaps I’m just not hard-core enough about this stuff—I don’t like high-CFM CPU coolers, either—but this thing isn’t for me. You will want to think long and hard before you decide that you can live with one of these cards in your PC.

Now, let’s get on with testing this thing in action.


Our testing methods
As ever, we did our best to deliver clean benchmark numbers. Tests were run at least twice, and the results were averaged.

Our test system was configured like so:

Processor Athlon XP ‘Thoroughbred’ 2600+ 2.083GHz
Front-side bus 333MHz (166MHz DDR)
Motherboard Asus A7N8X Deluxe
Chipset NVIDIA nForce2
North bridge nForce2 SPP
South bridge nForce2 MCP-T
Chipset drivers 2.03
Memory size 512MB (2 DIMMs)
Memory type Corsair XMS3200 PC2700 DDR SDRAM (333MHz)
Sound nForce2 APU
Storage Maxtor DiamondMax Plus D740X 7200RPM ATA/100 hard drive
OS Microsoft Windows XP Professional
OS updates Service Pack 1, DirectX 9.0

The test system’s Windows desktop was set at 1024×768 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

We used Catalyst revision 7.84 drivers for the Radeon 9800 Pro, and ATI cards and Catalyst 3.1 (7.83) drivers for the 9700 Pro. We used NVIDIA’s 42.68 drivers for the GeForce4 Ti 4600, and we used the brand-spankin’-new 43.45 drivers for the GeForce FX 5800 Ultra.

We used the following versions of our test applications:

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.


Fill rate
Even today, pixel-pushing power is still one of the key determinants of the performance of a graphics chip. Getting a handle on the GeForce FX 5800 Ultra’s fill rate performance, however, is more than a little slippery. Let’s start by looking at our chip table, and then I’ll tell you why it’s not exactly right.

  Core clock (MHz) Pixel pipelines  Peak fill rate (Mpixels/s) Texture units per pixel pipeline Peak fill rate (Mtexels/s) Memory clock (MHz) Memory bus width (bits) Peak memory bandwidth (GB/s)
GeForce4 Ti 4600


4 1200 2 2400 650 128 10.4
GeForce FX 5800 400 4 1600 2 3200 800 128 12.8
GeForce FX 5800 Ultra 500 4 2000 2 4000 1000 128 16.0
Radeon 9700 275 8 2200 1 2200 540 256 17.3
Parhelia-512 220 4 880 4 3520 550 256 17.6
Radeon 9700 Pro 325 8 2600 1 2600 620 256 19.8
Radeon 9800 Pro 380 8 3040 1 3040 680 256 21.8

The GeForce FX 5800 Ultra runs at much higher clock rates than its competition, and its DDR-II memory does, too. NVIDIA chose a cutting-edge approach to developing the GeForce FX, relying on newer technologies and higher clock rates to deliver performance. The chip’s 500MHz core clock speed gives it relatively high pixel and texel fill rates. The NV30 has four independent memory controllers in a crossbar arrangement, which is essentially the same as the GeForce3 and GeForce4 Ti chips, with the exception that the NV30 been tweaked to support DDR-II-style signaling. Its memory bus is only 128 bits wide, but 1GHz DDR-II memory gives the GXFX 5800 Ultra memory throughput of 16GB/s.

ATI has taken different approach with the Radeon 9700 and 9800 series, settling for lower clock rates but getting more work done each clock cycle. The high-end ATI chips have 256-bit-wide memory interfaces (with four 64-bit memory controllers), which give them more memory bandwidth than the GeForce FX 5800 cards, even with conventional DDR memory and the concomitant lower memory clock speeds.

Now, here’s why the above isn’t quite right. Our usual assumptions about graphics chips pipelines don’t entirely apply to the GeForce FX. NVIDIA is very coy about exactly how the NV30 GPU looks inside. For quite a while, most of the world believed NV30 was an 8-pipeline design with one texture unit per pipe. Turns out that isn’t so. Instead, the FX 5800 Ultra is… well, complicated. Lately, the company’s representatives have taken to talking about arrays of functional units instead of pixel pipelines. It’s sometimes hard to penetrate.

When asked, NVIDIA explains the NV30’s capabilities like so:

It renders:

8 z pixels per clock
8 stencil ops per clock
8 textures per clock
8 shader ops per clock
4 color + z pixels per clock with 4x multisampling enabled

It is architected to perform those functions.

Basically, its 8 pipes with the exception of color blenders for traditional ROP operations, for which it has hardware to do 4 pixels per clock for color & Z. It is that it has 8 “full” pipes that can blend 4 pixels per clock with color.

Now, the phrase “color + Z pixels” in there is key for our discussion, because that’s generally the kind of pixels most current 3D applications are rendering. That’s your standard pixel, with a color value, situated in 3D space. When doing this sort of conventional rendering, the NV30 can produce four pixels per clock cycle with up to two textures applied to each.

This configuration gives the NV30 a bit of a disadvantage next to ATI’s R3x0 series in terms of single-textured fill rate. Our table above reflects that difference, and it’s generally correct so far.

However, the NV30 can do certain types of operations, including stencil ops, at 8 pixels per clock. This ability makes the NV30 more formidable than a straight-up “4 x 2” pixel pipeline specification might indicate. NVIDIA claims the rendering and shadowing techniques used in upcoming games like Doom III will take particular advantage of the NV30’s eight-pipe capabilities.

Of course, we can measure these things. Here are the scores from 3DMark2001’s fill rate test, which is a simple test of traditional “color + Z” rendering.

The GeForce FX turns in performance more or less like we’d expect given the numbers on our chip table above, very much like a “4 x 2” design. Its single-textured fill rate is lower than the Radeons, but its multi-textured fill rate is second to none.

These numbers bode well for the GeForce FX in scenarios where games and applications apply at least two textures per pixel. However, some newer games are substituting pixel shader effects for additional textures, so the underlying strength of ATI’s true eight-pipe approach remains formidable.

In situations where applications want to apply more than one or two textures to a pixel, both the NV30 and the R300 series chips can “loop back” pixels through their pipelines in order to apply more textures. Both chips can apply up to 16 textures without resorting to multipass rendering.

The primary constraint for fill rate is sometimes memory bandwidth rather than GPU processing power, and memory bandwidth bottlenecks are more of a concern than ever when dealing with 64-bit and 128-bit floating-point color modes. In cases where memory bandwidth is a primary limitation, the Radeon 9700 and 9800 cards will have the advantage.

Occlusion detection
Then again, modern GPUs employ a whole range of tricks to make better use of their fill rate and memory bandwidth, including compression of various types of graphics data. Both the NV30 and its R3x0-series competition can compress Z data (depth information on pixels) and color (framebuffer) data using lossless algorithms. Color compression is new in this generation of chips, and it’s most useful when edge antialiasing techniques are active, where multiple samples are often the same color. As I understand it, the NV30’s color compression is always active, while ATI only turns on color compression with antialiasing.

NVIDIA doesn’t talk much about it, but I believe the NV30 employs an “Early Z” occlusion detection algorithm, like the R3x0 chips, to reduce the possibility the chip will render a pixel that would be situated behind another pixel—and thus not visible—in the final scene. With fancy shader programs in the mix, pixels become more expensive to render, so eliminating occluded pixels up front becomes a higher priority.

All of these methods of bandwidth conservation improve efficiency, and if implemented well, they offer the NV30 some hope of outperforming the ATI chips, even with less memory bandwidth. We’ll use VillageMark to test these chips’ fill rate and occlusion detection capabilities, both with and without antialiasing enabled.

Even with less memory bandwidth, the GFFX 5800 Ultra matches up well against the ATI cards, leading the pack in the non-AA tests. The ATI cards lead when 4X edge antialiasing and 8X anisotropic filtering are enabled, but even then, the GeForce FX runs remarkably close to the Radeon 9700.


Pixel shaders
The NV30’s pixel shaders can execute shader programs with as many as 1024 instructions in a single rendering pass with dynamic branching and looping. This ability should allow the NV30 to handle complex shader effects with grace. The chip’s limits are quite a bit higher than the 64-instruction limit imposed by DirectX 9’s pixel shader 2.0 specification.

By contrast, the Radeon 9700 Pro conforms largely to the PS 2.0 specification; it can execute a maximum 64 instructions per pass, with few exceptions. More complex shader programs will only be possible on the 9700 with the aid of a high-level shading language, which can break down effects into multiple rendering passes. The overhead associated with multiple rendering passes can harm performance, as well.

ATI addressed this limitation in the R350 chip by adding an “F-buffer” that stores intermediate pixel fragment values between passes through the pixel shaders; no trip through the rest of the graphics pipeline is required. With the F-buffer, the R350 can execute pixel shader programs of any length. You can read more about it in our Radeon 9800 Pro review.

We’ll have to see whether NVIDIA’s approach is superior to ATI’s when we have tests available to us that compile to our test hardware from high-level shading languages. Currently, the benchmarks available to us are limited to basic DX8 and DX9 shader variants, from 1.1 to 2.0.

Differences between ATI’s and NVIDIA’s pixel shader implementations add more complexity to the task of comparing the two companies’ chips. The NV30’s pixel shaders are composed of an array of arithmetic logic units, and the bit depth of pixel shader data affects clock-for-clock performance. In order to balance performance versus precision, NV30 offers support for two floating-point color bit depths, 16 bits per color channel (or 64 bits total) and 32 bits per color channel (or 128 bits total). ATI split the difference between the two; the R3x0-series pixel shaders process data at 24 bits per color channel, or 96 bits total, even when using 128-bit framebuffer modes.

NVIDIA’s approach offers developers more flexibility. They can choose higher-precision datatypes when needed, and they can fall back to 64-bit color when it is sufficient. However, ATI’s 96-bit compromise isn’t necessarily a bad one for this first generation of chips with high-color capabilities. The 96-bit limitation will probably be more of a disadvantage in professional rendering applications where the R3x0 chips will live on FireGL cards instead of Radeons.

I should mention one more thing about the GeForce FX’s pixel shaders. All of this funky talk about “arrays of computational units” got me wondering whether the FX architecture doesn’t share computational resources between its vertex and pixel shaders, which is likely to happen as graphics hardware evolves. So I asked NVIDIA, and the answer was straightforward:

The GeForce FX architecture uses separate, dedicated computation units for vertex shading versus pixel shading. The benefit is that there is never a trade-off between vertex horsepower and pixel horsepower.

In future GPUs, pixel and vertex shader instruction sets are likely to merge, and the walls between these two units may begin to dissolve. This change may enable some impressive new vertex-related effects. However, we’re not there yet.

Now, let’s test what we can, which is DX8 and early DX9 pixel shader performance.

Despite its advantage in clock speed, the GeForce FX comes out behind the Radeon 9800 Pro in all of our DirectX 8-class pixel shader tests. The FX decisively outperforms the 9700 Pro in one benchmark, 3DMark 2001’s pixel shader test, but otherwise runs behind both ATI chips.

Our sole DirectX 9-class shader test is 3DMark03, and here we come to a complication. NVIDIA has apparently optimized its 43.45 drivers specifically for 3DMark03, possibly by cutting pixel shader precision from 128 to 64 bits—or maybe less in some cases. (NVIDIA has given itself some cover on this front by raising objections to 3DMark03’s testing methodology.) To illustrate the performance difference, I’ve also tested the GeForce FX with revision 43.00 drivers, which don’t include the 3DMark03-specific optimizations. Both sets of results are presented below.

3DMark03’s pixel shader 2.0 test creates procedural volumetric textures of wood and marble via pixel shader programs. This is but one use of pixel shaders, but it may be common in future apps and games.

With the 43.45 drivers, the GeForce FX is very competitive with the ATI cards. With the 43.00 drivers, the GeForce FX can’t keep up. We’ll explore the issue of 3DMark03 optimizations in more detail below.

From what we can tell, though, the GeForce FX isn’t significantly more powerful than the ATI chips when running DirectX 8 or early DX9-class pixel shader programs. In NVIDIA’s own ChameleonMark, it’s consistently slower than the ATI cards, in fact.

Vertex shaders
If you were expecting me to tell you the number of vertex shader units the GeForce FX 5800 Ultra has, you probably haven’t been paying attention. ATI has laid it out clearly: the R300 and R350 each have four vertex shader units, each consisting of a 128-bit vector processor and a 32-bit scalar processor. The GeForce FX 5800 Ultra has—you guessed it—an “array” of computational units.

Ah, the disclosure.

Anyhow, finding out the relevant info about the FX isn’t too difficult. We’ll test the cards and see how they perform.

The ATI chips’ vertex shaders appear to be more powerful than the NV30, despite the fact the NV30 runs at a higher clock speed. The FX benefits from specific optimizations in 3DMark03, but not quite enough to catch up with the 9700.

Now let’s test legacy transformation and lighting performance, which is generally provided by a vertex program on newer cards.

The FX excels here, probably because old-school T&L runs on a single vertex shader unit, and the FX’s higher clock speed becomes more of an asset.


Quake III Arena
Now for the gaming tests. We tested with a Quake III demo from a CPL match involving my bud fatal1ty, of course. It’s a longish demo for benchmarking, but it should be a nice test. You can grab the demo from our server here, at least until we find out the thing is copyrighted somehow.

The GeForce FX 5800 Ultra performs very well in Quake III, outpacing the competition ever more decisively as the display resolution increases. With anisotropic filtering and edge AA enabled, the FX’s advantage grows slimmer, though.

Comanche 4

Comanche 4 is obviously limited by our test platform rather than by graphics cards when AA and aniso aren’t enabled. However, once we turn up AA and aniso, the GeForce FX 5800 Ultra runs neck-and-neck with the Radeon 9700 Pro. The 9800 Pro, though, is faster at higher resolutions.

Codecreatures Benchmark Pro
The Codecreatures benchmark makes extensive use of DirectX pixel shaders, so it’s an intriguing test for these cards.

Here we see the opposite result of what we saw in Quake III. The Radeon 9800 Pro is fastest without aniso and AA, but the FX pulls way out into the lead once we turn up the eye candy.


Unreal Tournament 2003
Here we have the latest and greatest multiplayer first-person shooter engine. UT2003 uses larger textures and more polygons than older games, and it uses pixel shaders for some effects.

Without aniso and AA, the FX battles the ATI cards for supremacy. The FX does relatively better with UT’s high detail settings than with lower detail.

With AA and aniso, the pattern reverses itself. The FX is competitive at the low-detail settings, but has trouble keeping up at the higher detail level.


Serious Sam SE
In order to test Serious Sam on a more-or-less level playing field, we used Serious Sam’s “Extreme Quality” add-on to set graphics options. For the most part, that will mean the graphics settings are comparable from card to card, with one exception. The Radeon cards are doing 16X anisotropic filtering here, and the NVIDIA cards are at their maximums of 8X aniso. However, with the adaptive aniso algorithm that ATI uses, the difference between 8X aniso and 16X aniso is very minor.

The GeForce FX blows away the Radeon cards here, averaging about 20 fps faster at everything but the lowest resolution.

Serious Sam lets us look at actual performance over time rather than just averages, so let’s see what we can see. One thing to look for: performance dips. In many ways, the quality of the gaming experience is defined by what happens in the worst-case scenarios.

The GeForce FX’s dominance isn’t adequately captured by its average scores. At 1600×1200, the FX is so fast, its performance low-points are higher than the previous-gen GeForce4’s performance peaks.

With 4X AA plus 8X aniso, the FX’s performance lead narrows, but remains intact. Let’s look at the performance over time.

Again, the FX shows impressive strength across the board.


3DMark2001 SE

The FX splits the difference with the Radeon 9700 Pro, losing the two lower resolution tests, but making up ground as fill rate requirements rise. The Radeon 9800 Pro, however, is fastest overall.

You can see how the cards perform here in each of the individual 3DMark game tests. The most interesting test is the Nature game, where DX8 pixel shaders play a prominent role. The FX is sandwiched neatly between the Radeon 9700 Pro and 9800 Pro in the Nature test.


Now we’re back to 3DMark03, where we need two sets of drivers in order to test adequately. NVIDIA’s optimizations in the 43.45 drivers affect not only performance, but image quality. I’ll show you what I mean, but let’s look at the performance results first.

The GeForce FX 5800 Ultra is outright fastest in 3DMark03 when using the 43.45 driver, but it’s much slower with 43.00. The individual game test scores make up this composite “3DMark03” score. The results of those individual tests follow.

Obviously, NVIDIA has made considerable performance progress through driver optimizations, particularly in the Game 4 test, Mother Nature, which uses Direct X 9-class pixel shaders. Let me show you why that is.


3DMark03 image quality and driver optimizations
There are several screenshots below taken from frame 1,799 of 3DMark03’s Mother Nature test. These images are fairly low-compression JPEGs, but you can click them to see full-screen versions of the images in lossless PNG format. I’ve performed a gamma correct 1.1 on the JPEG versions to make things more visible in them. The PNG versions are unaltered.

You’ll want to notice several things about the images. The first image, for instance, was produced by the DirectX 9 reference rasterizer. It’s a software-only tool that takes hours to produce this single image, which is useful for comparison’s sake. The next image comes from the Radeon 9800 Pro, and it looks very similar to the output from the reference rasterizer.

The next two images come from the GeForce FX with two different driver revisions: 43.00, which hasn’t been optimized for 3DMark03, and 43.45, which has. The 43.00 image has some corruption on the rock’s edge, right beneath the flowing stream. Also, the butterfly’s shadow is too large and exhibits artifacts in the 43.00 image. 43.45 corrects these problems, but creates a new problem of its own: the dynamic range on the sky simply isn’t what it should be. Somehow, NVIDIA has cut the precision down enough in the process of optimization that the scene doesn’t look as it should.

I asked NVIDIA how they’d optimized for 3DMark03 in their newer drivers, but they didn’t answer me. Had they simply cut all pixel shader precision to 64 bits in floating point, I wouldn’t have objected too strenuously. A case could be made for using 64-bit FP datatypes for this sort of performance test. However, whatever’s going on with the sky in the 43.45 drivers looks like it goes well beyond such a reasonable adjustment.

Direct X 9 reference rasterizer

Radeon 9800 Pro

GeForce FX 5800 Ultra with 43.00 drivers

GeForce FX 5800 Ultra with 43.45 drivers


Next up is antialiasing performance, which we’ve been testing in most of our game tests by turning up 4X AA and 8X aniso. Now we’ll isolate the various AA and texture filtering modes to see how the cards scale with each of them.

Edge antialiasing
These AA tests are intended to show performance scaling, but they are not entirely comprehensive. The GeForce FX includes a broad range of edge antialiasing modes, and we didn’t test all of them. For instance, we skipped NVIDIA’s quirky “Quincunx” mode, which combines 2X AA with a blurring filter. I find that mode essentially useless.

Also, NVIDIA has a couple of AA modes called “4XS” and “6XS” that combine multisampled edge AA with full-screen supersampling. When I tested “4X” AA, I used the usual multisampled mode. Since the FX doesn’t have a straight-up 6X multisampled mode, I compared the FX’s 6XS mode to the Radeon cards in their 6X mode.

Finally, the 43.45 drivers have a slider option for 16X AA, but I found no performance difference between 8X and 16X mode at all. I didn’t include those results, because I couldn’t be sure the 16X mode was functioning properly.

The FX takes a relatively big hit when going from 4X mode to 6XS mode, which is to be expected, since supersampling is much less efficient than multisampling. All in all, though, the FX performs pretty well with edge AA in use. The FX scales much better than the GeForce4 Ti, and it’s comparable to the ATI chips.

There’s much more to say about these chips’ edge antialiasing methods, but I will have to try to address those in more detail in another article. For now, we need to concentrate on another type of antialiasing: texture filtering.


Texture antialiasing
To test texture filtering, I used Quake III Arena’s timedemo function at 1600×1200 resolution with various filtering settings.

The GeForce FX runs Quake III especially well, so it has a bit of advantage here in overall performance. In terms of scaling, there aren’t many surprises, either. The FX takes a little more of a performance hit with trilinear filtering than the GeForce4 does, but it’s so much faster overall, no one will complain.

Incidentally, I used the “Quality” setting in the NVIDIA “Performance & Quality Settings” control panel throughout my performance testing. There are two other options, “Application” and “Performance”. Let’s take a look at what the three settings do.

Below are some low-compression JPEG images from Quake III that show the effects of the various filtering settings. You can click the images to see full-quality PNG versions. I’ve arranged the screenshots from highest quality filtering to lowest, and the difference is very visible in the look of the textures in the scene, especially in the long bridge surface stretching out in front of the viewpoint.

GeForce FX “Application” option: 8X aniso + trilinear filtering

GeForce FX “Quality” option: 8X aniso + trilinear filtering

The difference between “Application” and “Quality” is difficult to see with the naked eye, but you can tell by looking at the definition of the seams between the tiles underneath the railgun, if you look closely. (The railgun is in the player’s weapon sights, floating above that little platform there.) The definition of textures is cleaner with “Application,” even right up next to the viewpoint.

GeForce FX “Performance” option: 8X aniso + trilinear filtering

I’m not sure how NVIDIA can call the “Performance” option 8X anisotropic filtering. The number of samples here is obviously lower, and there’s visible banding along the Z axis as the eye scans up the platform. The tile seams under the railgun are no longer distinct.

GeForce FX Standard isotropic filtering (bilinear + trilinear)

I included this sample of trilinear + bilinear filtering to show you how little difference there is between this filtering mode and the “8X” “Performance” option. Compare this image to the one at the top of the group to see how much difference anisotropic filtering can really make.

Here are some images from the Radeon 9800 Pro’s various filtering modes for comparison. ATI has clearly delineated the difference between its “Performance” and “Quality” options: in “Performance” mode, the driver doesn’t allow trilinear filtering to happen in conjunction with aniso, even if the application requests it. “Quality” mode simply obliges apps when they ask for trilinear. You can see the difference in the mip-map transitions in the images below, but it’s even easier to see in action, when you get those annoying “lines” moving around in front of you as you move through a game.

Radeon 9800 “Quality” option: 8X aniso + trilinear filtering

Radeon 9800 “Performance” option: 8X aniso + (forced) bilinear filtering

Radeon 9800 standard isotropic filtering (bilinear + trilinear)

Fair warning: if you’re on dial-up, I recommend skipping the next two pages. I’ve embedded PNG images into the page because JPEG Simply Won’t Do, and you dial-up users would do better to just jump past those pages.

Now let’s see exactly what NVIDIA is doing to texture filtering with these various quality slider settings. I’ve used Q3A’s “r_colormiplevels 1” command to expose the various mip-maps in use and the transitions between them.

GeForce FX “Application” option: 8X aniso + trilinear filtering

GeForce FX “Quality” option: 8X aniso + trilinear filtering

GeForce FX “Performance” option: 8X aniso + trilinear filtering

GeForce FX Standard isotropic filtering (bilinear + trilinear)

The “Application” setting looks gorgeous, with long, smooth gradients between a small number of mip-map levels. The “Quality” option isn’t quite as pretty, and the texture level of detail is a little lower here. Gradients are still smooth, but less so.

The “Performance” option looks very similar to the trilinear + bilinear option, but remarkably, the transitions between mip maps are very sudden. That explains the banding along the Z axis that we saw in our previous set of example shots.

NVIDIA’s “Performance” mode is an interesting beast. Perhaps some folks will find this combination of filtering methods useful if they have GeForce FX 5200 and 5600 cards, but I have a hard time imagining why a GeForce FX 5800 Ultra owner would choose to compromise his card’s image quality in this way.

Here are Radeon 9800 screenshots for comparison. Although it doesn’t in every case, in this example the ATI “Quality” setting looks nearly as good as NVIDIA’s “Application” mode. You can see how the “Performance” option simply turns off trilinear filtering without altering anything else.

Radeon 9800 “Quality” option: 8X aniso + trilinear filtering

Radeon 9800 “Performance” option: 8X aniso + trilinear filtering

Radeon 9800 Standard isotropic filtering (bilinear + trilinear)
You’ve probably seen enough of the GeForce FX 5800 Ultra now to draw some of your own conclusions. For me, wrapping up this review is a little odd, because it feels more like a post-mortem than a new product review. NVIDIA won’t confirm it, but the near-universally-accepted rumor is that very few NV30-based products, either the GeForce FX 5800 Ultra or the plain-jane GeForce FX 5800, will ever see store shelves in North America. NVIDIA’s add-in board partners haven’t been receiving supplies of NV30-based cards in any kind of volume, and none I’ve talked to are optimistic about ever receiving them.

Meanwhile, rumors are flying that the follow-on to NV30, the NV35, is up and running in NVIDIA’s labs and coming to market fairly soon. NVIDIA has formed a new manufacturing partnership with IBM, whose semiconductor fabrication technology is top-notch. The NV35 is expected to outperform the current Radeon 9800 Pro without the need for a Dustbuster cooling system.

But I do have a few definite opinions about the GeForce FX 5800 Ultra before it rides off to take its place alongside the 3dfx Voodoo 5 6000 in the Museum of Large Video Cards That Didn’t Quite Make It. This would have been a great product had it arrived six months earlier at this same clock speed with lower heat levels, a more reasonable cooler, and lower prices. As it stands, the GeForce FX 5800 Ultra is not a good deal, and I wouldn’t recommend buying one. Yes, it’s very fast, especially in current games. It’s also loud, expensive, and did I mention loud?

Go get a Radeon 9800 Pro if you want a high-end graphics card. As time passes, I continue to marvel at how much ATI got right with the R300 chip and its derivatives. It simply has very few weaknesses. The technology is sound, and the choices they made in building the product were very smart.

Also, I’m not compelled by NVIDIA’s talk of “arrays of computational units” and the like to describe the NV30. I’m not convinced they’ve changed their way of designing GPUs; I think they’ve mainly just changed their way of talking about them. In doing so, NVIDIA has managed to obscure some basic information about the chip. The NV30 doesn’t offer any uniquely compelling technological advantages over the R300-series GPUs, judging by what we’ve seen, and I’m still waiting to find out how the NV30’s purportedly different design offers capabilities or flexibility unmatched by its competitors.

The NV30’s basic technology will live on across NVIDIA’s product line, from the upcoming NV35 to the dirt-cheap GeForce FX 5200, all of which will have DirectX 9-class rendering abilities. As a foundation for NVIDIA’s lineup, the NV30 isn’t bad. We’re still waiting to test truly intensive next-generation software written with high-level shading languages to see how the NV30 handles those. Indications are, though, that the ATI chips have more powerful pixel and vertex shaders, not just on a clock-for-clock basis, but even with a pronounced clock-speed disadvantage. Whether NVIDIA can compete well with NV30-derived technology will depend on how good those particular implementations—NV31, NV34, and NV35—truly are. NV30 technology will give those chips rough feature parity with ATI, but performance is the question. 

Comments closed
    • Krogoth
    • 15 years ago

    5800U was an overambitious and overhyped part that did not follow the DX9 spec to the letter. It tried to use new, hot from the factory (pun intended) DDR2. The result is a die-shrink, massive clocked Ti4x00 with psudo DX9 support and dumb driver tricks so their penismark scores would match their ATI counterparts. Nvidia eventually recovered from this mishap with the NV40.

      • bthylafh
      • 13 years ago

      Holy thread necrophilia, Batman! Are you that sad and uninteresting that you had to post to the thread three years after the last post?

      {troll, troll}

    • Pete
    • 18 years ago

    It’s also interesting/disturbing to note that nV’s AA actually _removes_ some clouds.

    • Pete
    • 18 years ago

    Here we go: §[<http://www.vr-zone.com/reviews/NVIDIA/FX5800U/page4.htm<]§ Obviously, it's tough to tell from such tiny shots, but the wings of the foreground plane look smoother in the ATi pic, probably due to gamma-correction.

    • Pete
    • 18 years ago

    Nope, I’ve seen more than a few pics showing 16x AA. (A recent [unexceptional] FX:U review, within the past week IIRC, showed a 16x nV pic next to a 6x ATi pic, both of 3DM03 GT1, and ATi’s looked better–likely because of gamma correction, as an ATi driver dev pointed out to me.)

      • Ardrid
      • 18 years ago

      Ahh…they added 16X under the 43.45 Dets.

    • Pete
    • 18 years ago

    (Lump me in the category of ppl who nod at TR’s reviews.)

    • Anonymous
    • 18 years ago

    /me just simply shakes his head at tech-report…

      • Anonymous
      • 18 years ago

      /Me just nods at Tech-Report
      Everybody is free to consider this review as useful or not.

    • quarantined
    • 18 years ago

    I think #11 is right about the image quality tests in 3DMark03. The 43.00 drivers show frame 1799 and all the others are frame 1779, according to the 3DMark03 table at the bottom of each screen. After alt+tabbing between these it’s obvious that it’s not the same frame. Or am I missing something here?

    Anyways, good review.

      • Damage
      • 18 years ago

      Doh! Fixed.

    • Anonymous
    • 18 years ago

    Wow, downhill spiral. Did you really take my comments with so much hostility that you felt such hostility in return was in order? Which comments among the “you have a point” and “BTW, if I’d disagreed with the people who said it was a good review (other than this glaring and extremely common problem), I’d have mentioned it” gave you the impression of personal attack?

    I tried to use bold to emphasize, though you seem to take it as an offense. I’m sorry, I wasn’t aware it would appear so thick and bright as to appear to be shouting.
    Your taking the opportunity to make a mocking commentary was disappointing.

    I was offering criticism where I felt it was warranted.

    I had thought you analyzed the mip map boundaries in motion. If not, the observations that state Balanced/Quality mode mip map issues remain informal observations (like this: §[<http://www.beyond3d.com/forum/viewtopic.php?t=4772<]§ ), whereas I thought a review had finally discussed and contrasted the subject in detail. My mistake. Some further comments: I am already aware that mip map boundaries are hard to see in screenshots. Beyond3D does indeed say that the nv30 only does Trilinear with Application mode. They discuss their nv30 article findings and the actions they are taking for their recent review here: §[<http://www.beyond3d.com/previews/nvidia/nv31/index.php?p=4<]§ This issue wasn't one of being "harder on the GF FX", but of misrepresentation. Since most sites do exactly what you have done, benchmark the defaults as provided by the drivers for comparison as if they are equivalent because they are the defaults, or for being labelled the same, it does seem to me to be a consistent and significant issue. It also seems to me an issue directly intended by nVidia by offering Trilinear only in "Application" mode, and renaming Balanced and Aggressive as they intend them to be benchmarked against the 9700's settings. Therefore, I do view them "getting away" with that nearly universally as significant. I understand that you disagree, but I do not think my opinion warrants disparagment, nor do I think I'm magnifying something insignificant to a "monumental" degree as it is giving a tangible increase to GF FX benchmark figures, and it does lend itself to expectations of performance being "fixed" to match the higher results for other applications. On a parting note, I did not say you thought you were flawless...if I thought you'd believed that, why would I have posted? I am glad that aside from the verbal sparring you did take my input on board, and I just wish criticism of this issue could have been taken other than personally.

    • Anonymous
    • 18 years ago

    “Other NVIDIA partners have cooked up alternative cooler designs for the FX 5800 Ultra, but most cards will probably be similar to this one”
    Well i have a Leadteak 5800 Ultra, and it’s cooler solution is not the same,
    no annoying sound on my card, and only takes one pci slot.

      • indeego
      • 18 years ago

      You mean AGP, right? Or is it considered a good thing that the Gforce line is /[

        • Anonymous
        • 18 years ago

        I ment AGP, but i guess a card in the first pci would melt, lol

    • Dposcorp
    • 18 years ago

    Excellent review. Worth the wait.
    Best line it? None, its in the comments. (#46 to be exact).


    • Anonymous
    • 18 years ago

    So basically Damage spent $560 some odd dollars on a vid card everyone knew was too little, too late.

    Damn shame that Nvidia screwed up with the NV30. But I just hope ATI and Nvidia and Matrox and BitBoys Oy will continue to push the designs harder and faster so it benifits you, me, and my left big toe. And damnit… they better start getting cheaper. I refuse to buy a vid card that’s $400+ that’s loud and annoying. At that price they better include some midget to get in my computer and blow on the card to keep it cool.

    • Anonymous
    • 18 years ago

    This is AG #15 (who is not AG #27)

    All the reader can tell is that there is some mip map blending going on. While that’s prettier than no blending *[

      • Damage
      • 18 years ago

      Bold letters: the key to everything. Who knew?

      I find myself having to say this a lot, but it’s sadly true too often: Your tone could use some adjustment. I don’t see the point in being so confrontational just because you disagree.

      You seem to have several arguments with me. One is the fact I didn’t address these things in enough detail in the review, which I’ve tried to rectify by responding here–earning your kindness in the process. 🙂 I’m sorry I can’t be infallible. I never claimed to be the pope of graphics reviews. We do try to answer questions and respond to feedback, though.

      I don’t believe I said anything about trilinear (w/o aniso) mip map boundaries in motion on the NV30 vs. the R350. Not sure where you got that.

      The massive visual difference of which you speak between the NV30 in “Quality” mode and the R350 in “Quality” mode with bilinear + trilinear looks like this:

      §[<http://tech-report.com/reviews/2003q2/geforcefx-5800ultra/q3a-trilinear-fx.jpg<]§ versus this: §[<http://tech-report.com/reviews/2003q2/geforcefx-5800ultra/q3a-trilinear-9800.jpg<]§ I'm sorry, but I just don't see it, and I *enjoy* being picky about such things when appropriate. I'm glad reviewers and consumers are getting more sophisticated about this stuff, but we've seen much bigger visual differences between cards in comparison too many times to count. NVIDIA may be getting away with something here, but it ain't much. Sometimes, performance tweaks that make little visual difference are appropriate and even welcome. We are considering how to handle this issue in future reviews, and we may be persuaded to take the approach you advocate. I never said I had the One Right Answer to these issues. However, I am not at all persuaded that the difference between is as monumental as you make it out to be. Not even close. As for the aniso issue, I like your suggestion (testing 8X versus 16X). Both approaches are coping methods, though, and not exact methods of comparison.

        • Anonymous
        • 18 years ago

        Well, nvidia/ati’s drivers (or rather, the settings you are using) are not doing as the application directed, which make your benchmarks flawed, since you did not specify in the review what the image qualities are, and how much it impacts the numbers you are posting. So given this consideration, should we post benchmarks at an arbitrary point of (in)comparable image quality, or get the highest quality a card can produce and match it to the other card?

        Come to think of it, what are the driver settings you are using? Because I’m not sure if nvidia has a “quality” mode.

        Also, I doubt if anybody can find any differences on the jpegs you linked, since they are jpegs, and 1/4 of the resolution of they are captured at.

    • Disco
    • 18 years ago

    unrelated question – how do you get the little ‘smilies’ to show up? Are they automatic?

    • crazybus
    • 18 years ago

    I can see what the point of nvidia’s performance AF mode is. As you can see in the screenies the LOD is not any better with 8xAF on than with no AF and better filtering. I have a Radeon 8500 and I know that many games don’t look too good without trilinear filtering with the AF.

      • crazybus
      • 18 years ago

      oops……….I can’t see

    • Pete
    • 18 years ago

    Damage, your review was, as usual, a pleasure to read. I particularly liked the extensive screenshot comparisons. (The Serious Sam numbers are very interesting, and I wonder if that game is very dependent on vertex shaders, as the Radeon 9000 Pro, with its shader borrowed from R300, seems to perform much better than the 8500 on SS. Or perhaps the 9000 is an anomaly, shader performance isn’t that important, and SS just likes a 500MHz 4×2 card better than a 380MHz 8×1, which is very reasonable.) I have a few questions, though, particularly since nVidia so greatly stressed the “Cinematic” aspect of their new architecture:

    I don’t think the FX is using FP16 in 3DM03, I think they’re using INT. B3D had a recent thread where an FX user posted test numbers, and there was no difference between FP16 and FP32, whereas using INT (12-bit) resulted in a 3-4x boost.

    I guess you checked out the MIP-maps in Q3 very well, as evident by your screenshots, but did you verify equal LOD? I’m still troubled by mboeller’s post here: §[<http://www.beyond3d.com/forum/viewtopic.php?t=5074&postdays=0&postorder=asc&highlight=quack&start=20<]§ Pity I can't read German, so I can't fully understand what the author is writing, but the pictures seem clear. nV seems to have lowered the quality of the LOD (blurrier textures), thus boosting their framerate. Something seems fishy, but I don't know if it's the site or the drivers in question. I think you should have noted that nVidia switched their AF terminology from App/Bal/Aggr in the 42.xx's to App/Qual/Perf in the 43.45, specifically to invite comparisons to ATi's Qual/Perf. You note that ATi's Qual uses trilinear filtering, and you note that nV's "Qual" doesn't have MIP-map transitions as smooth as tri, but you don't seem to state clearly that nV isn't performing tri when it uses Bal/"Qual"--it's using a mix of bi and tri (meaning, only rough transitions at the intersection of MIP-map boundaries, not smooth transitions across the whole MIP-map as with tri). True, there are other trade-offs with ATi's AF, in that it's not full-sample at all angles, which may be important with flight-sims; but for the majority of FPS's, ATi should offer superior sampling on most surfaces. It's a toss-up, I think, but the bi vs. tri issue is important. 3DVelocity's review also showed how nV's AF is pretty terrible in certain racing sims, so there may be more problems on non-standard benchmark games. While we're on the topic of IQ, I think you were remiss not to delve deeper into the differences between nV's and ATi's AA. ATi's is clearly superior because it uses a rotated-grid and employs gamma correction, whereas nV's "IntelliSample" (sounds like a sea of computing units, to me :P ) uses ordered-grid for their standard modes (read: the ones users will unknowingly compare directly to ATi's) and no gamma correction at all. You really should have used nV's 4xS, which is rotated-grid, to compare to ATi's 4x. Anyway, those are clearly more in-depth, investigative issues, and ones I hope you look into once the rumored upcoming WHQL Det 50's are released in a few weeks. Wring all you can out of that $600 museum piece! ;)

      • Damage
      • 18 years ago


      See here:

      §[<http://tech-report.com/ja.zz?id=2891<]§ And yes, more investigation into AA would be nice to do. I think the short answer is that gamma correction gives ATI superiority, and not by a little bit. In gamma-correct blends, we may finally have found something that matters nearly as much as sample size to overall AA quality. I'm not buying your 4XS vs. ATI 4X thing, though. :)

        • Pete
        • 18 years ago

        Damage, I still think of you as the pope of video card reviews.

        But not because you’re infallible–I just like the funny hats you wear. Your “Sir Graphs-a-Lot,” “ha-ha-rimshot,” and “witty captions” hats, mainly. 😉

        Yep, I read your huge reply above my equally-voluminous one after I posted (I got distracted), but my issue is with how the boundaries look in motion. I’ve read a review that stated nV’s Balanced blending still shows visible MIP-map lines in motion, so I was mainly wondering if you noticed that, too. Still, IMO the bottom-left and top-right columns as well as the floor show some clear differences between nV’s Balanced tri and ATi’s Quality’s tri. No big deal–some reviewers, like Hardware.fr, argue that nV’s Balanced is closer to ATi’s Quality in terms of overall compromises, while others, like [H], argue that only Application is an aples-to-apples comparison. I’m torn, but with FPS’s, I tend to side with the App crowd; with flight simulators and other 360-degree apps, I may differ. Digit-Life (or was it xbit-labs) continue to use Serious Sam to show ATi’s weakness at certain angles, and they’re right to do so; OTOH, I see reviews like 3DVelocity’s which shows ATi’s AF is clearly (almost laughably) superior to nV’s at every level. As long as you present the evidence from which you drew your conclusions, I’m cool with it.

        And, obviously, us Anonymous and Nonymous Gerbils will feel free to disagree with your interpretation of the 3D Bible. God Bless America, or the Internet, or … Something. The Constitution. Whatever, stop confusing me with your silly hats!

        As for my nV 4xS vs. ATi 4x assertion, yo, check it:
        §[<http://www.beyond3d.com//reviews/ati/r350/index.php?p=20#sample<]§ §[<http://www.beyond3d.com/previews/nvidia/gffxu/index.php?p=17<]§ The pics explain why 4xS actually looks better than 6xS or 8xS (neither of which nV can do in OpenGL, I believe, while ATi goes all the way to a glorious 6x) in most screenshots I've seen on the web. Note that 4xS presents pretty much the same exact rotated sample pattern as ATi's 4x; it's nV's problem that they have to combine super- and multi-sampling to achieve that effect, not ATi's. nV can add all the letters it wants to the end of its AA implementations, but that can't substitute for gamma-correction, which makes even nV's new 16x AA mode look worse than ATi's 6x mode. A pretty incredible addition to IQ on ATi's part, IMO, and one I hope nV imitates soon. Anyway, I agree with your conclusion, and that's all that counts: that you continue to feed my delusion. Cheers!

          • Ardrid
          • 18 years ago

          Pete, I’m assuming you meant NVIDIA’s new 8X AA. Matrox is the only company out there currently using 16X AA 🙂

    • Anonymous
    • 18 years ago

    From page 5 of the article:
    “y[ NV30’s pixel shaders can execute shader programs with as many as 1024 instructions in a single rendering pass with dynamic branching and looping.y]

    The NV30 has no such dynamic branching and looping inside the pixel shader. IIRC, it can do at most static looping and has no branching capabilities at all except to render all possibilities and pick the correct one. Dynamic branching and looping is a major feature of upcoming PS3.0 hardware such as R400 and NV4x.

    • Division 3
    • 18 years ago

    IMO – The GFFx 5800 Ultra would not have stood a chances against the R97Pro if it was clock for clock.

    • Hockster
    • 18 years ago

    Nice review. I know Quake III is an old game, but I must say that I’m impressed with the GFFX’s performance on it. Overall though, I prefer the Radeon 9800 Pro by a long way.

    • Alex
    • 18 years ago

    Haha… Nice Steel.

    Good review.

    • Steel
    • 18 years ago

    Looks like Damage could use a pair of these:
    §[<http://www.bose.com/noise_reduction/personal/qc_headset/index.html<]§ (They actually work; I was able to try a pair in our noisy computer room this weekend.)

      • indeego
      • 18 years ago

      Buy some $10 pair from Home Depot, they work much better. Sure you look stupid with earmuffs on in the data room, but then again, people avoid youg{<.<}g

        • Steel
        • 18 years ago

        Yeah, but the neat trick with the Bose headset is you can still hear people talking. Sometimes more clearly than with them off. ($300 /[

          • indeego
          • 18 years ago

          For a $290 difference I’ll take the hard route and just remove them when people talkg{<.<}g

    • Anonymous
    • 18 years ago

    As AG15 said, from what Beyond3D have shown, ONLY the application setting uses trilinear filtering, and is one of the reasons why the GFX looks half decent in this review.

      • Damage
      • 18 years ago

      As you know, making apples-to-apples comparisons between graphics chips is very tricky business. There is no true and exact image quality comparison to be had most of the time, so one strives to get as close as possible within reason. I did consider the issue you’ve raised as I worked through my review, and I’m not sure I have the right answer, but I will make an argument for the comparison as I conducted it.

      First, your assertion that the only way for the FX to run with Trilinear filtering certainly isn’t correct; Beyond3D doesn’t even claim that. They claim the “Balanced/Quality” setting uses a mixture of bilinear and trilinear filtering, which is a far cry from zero trilinear.

      My sample screenshots of trilinear + bilinear were taken with the “Quality” setting on the FX, and the images output by the FX and the Radeon 9800 are strikingly similar:

      §[<http://tech-report.com/reviews/2003q2/geforcefx-5800ultra/q3a-colormip-trilinear.png<]§ §[<http://tech-report.com/reviews/2003q2/geforcefx-5800ultra/q3a-colormip-trilinear-9800.png<]§ If NVIDIA is "cheating" here, they have managed to do so without compromising image quality horribly. The difference is difficult to detect in straight screen shots. Even with color mip levels, the difference between the two isn't all that great. I included screenshots, of course, to let the reader decide for himself. That's without aniso enabled. With aniso enabled, the Radeon 9800 Pro definitely looks more like the FX in "Application" mode in my examples. I admitted as much. However, ATI's aniso filtering algorithm still doesn't perform aniso on surfaces at certain degrees of inclination, and I'm not sure exactly how to account for that. (The effect, unfortunately, isn't easily visible in my example shots from Q3A; that shot shows mostly clean, square angles.) One way of doing so was allowing NVIDIA to fudge a little on trilinear. I could have been harder on the GeForce FX, and perhaps I should have been. I'll consider using the "Application" setting in the future However, honestly, it was a bit of a judgement call, and I knew the FX could use the benefit of the doubt. The potential difference in benchmark scores didn't affect the outcome of the review. My recommendation is, for a whole range of reasons, to buy a Radeon 9800 Pro if you want a high-end graphics card. The fact that NVIDIA's driver tricks raise more questions than ATI's simply underscores what I was saying about the fundamental strength of the R300-series chips. Scott

        • Anonymous
        • 18 years ago

        The Balanced/Quality mode might offer some trilinear filtering, but there are noticable differences in image quality, and comparing such a mode with ATI doing full trilinear isn’t fair IMO.

        The anisotropic filtering done by ATI might be a similar case, but there’s one major difference: ATI still has the best IQ when using their adaptive aniso mode.

        Personally I hope you’ll use the Application setting in future reviews, but if you decide not to do that, I hope you’ll give a good explanation why and show that the modes used are similar in IQ.

    • Ruiner
    • 18 years ago

    From the conclusion:
    “/[<"This would have been a great product had it arrived six months earlier at this same clock speed with lower heat levels, a more reasonable cooler, and lower prices. "<]/ lol. if...if...if....ifff 'And if a frog had wings, he wouldn't bump his ass a'hoppin.....or my name aint Nathan Arizona!!'

    • Anonymous
    • 18 years ago

    One good thing (IMO) in the design of this card is the choice of using thermal management control. (Probably) Nvidia was forced to do this and is a shame that both Nvidia and Ati do not use this on a regular basis, at least on their high-end cards. And take into account that they already have the technology from their mobile parts.


    • just brew it!
    • 18 years ago


    • nonegatives
    • 18 years ago

    What would a 97/9800 do if it were overclocked to match the 500 MHz of the 5800? Yea, I know it’s like comparing XP’s to P4’s, but we do that anyway.

    • MagerValp
    • 18 years ago

    My Sapphire 9700 just seems like a better and better buy…

    • Anonymous
    • 18 years ago

    Good lord, Scott’s channeling Dan from Dan’s Data…

    • thorz
    • 18 years ago

    Great review Damage. As others have pointed already the result has not been a sorprise, but its always a delight to read this kind of articles. I am going to get a 9800 pro at the end of this month. They are coming on 25 april to almost all the online stores in Norway, the Sapphire ones. They have recieved good reiews on their 9700 pro models. The price is still hight, arround US$600 after the convertion, but what the heck! With MS letting us rest of DirectX releases in the next 2 years(?) (at least its what they say) I think its a safe buy. I have had 2 geforces in the last years and they only lasted 1 year each. That sucks.

    • R2P2
    • 18 years ago

    You guys need to start using AA on your graphs. Some of those lines are pretty jaggy. 🙂

    Nice review, though. Hope the Dustbuster didn’t hurt your hearing too much.

    • Anonymous
    • 18 years ago

    Why did you compare a mode on the GF FX using Bilinear filtering to a mode on the 9700 using trilinear filtering? Just because they are the defaults for the respective cards? Or am I wrong, and you used (ATI)Performance for the 9700? I didn’t notice you saying explicitly for the 9700 aniso discussion, and my understanding is that (ATI)Quality is the default for the 9700.

    For the GF FX, only “Application” mode allows trilinear filtering when requested…all other modes force a limited blend bilinear filtering mode (even with no aniso). This mode should be technically better than plain bilinear filtering, but your observations indicate that the theory isn’t noticeable in practice.

    The problem is that you posted comparative results between the cards, and labelled the GF FX as “winning” for some benchmarks, and ,made no mention of the issue of comparing trilinear to bilinear (again, assuming I understand the defaults correctly). I’d be curious for an image quality AND performance comparison with the 9700 in (ATI)Quality with “Application preference” set, and the GF FX in Application. Obviously, the GF FX will suffer performance-wise with aniso turned on, but with no aniso requested by the application the benchmark results would actually be comparing the cards doing almost identical workload. Or, maybe running the 9700 in (ATI) Performance against the GF FX in (nVidia)Quality, and highlighting any benefits you noticed from the GF FX limited blend bilinear.

    Either would be a comparison actually testing the cards relative strengths rather than testing how little the default settings demand from the cards.

    • ieya
    • 18 years ago

    yokem55, reading the text on that page suggests the pics are the right way around; Scott notes that the later drivers fixed a bug with the butterfly’s shadow, but lost out on things like the sky,

      • yokem55
      • 18 years ago

      Ooops….I guess I was a bit to quick to look over the pictures.

    • Anonymous
    • 18 years ago

    I’s still like ATI to be at 128 bit FP right now.

    • EasyRhino
    • 18 years ago

    1) I’ve heard rumor that the retail Dustbuster is slightly quieter than pre-release Dustbusters. Any idea if that’s true?

    2) Comanche still doesn’t seem like a very useful benchmark to me

    3) Interesting, Damage’s writeup made the FX look better than others I’ve seen. My guess is because most of the game tests were only at 4x AA instead of 8x like other sites have used. Not saying that’s good or bad, it’s just a new take on things. 🙂

    4) Damage, wanna try removing the Dustbuster, and time how long it takes the FX to melt itself? That would be pretty cool.


      • droopy1592
      • 18 years ago

      q[<3) Interesting, Damage's writeup made the FX look better than others I've seen. My guess is because most of the game tests were only at 4x AA instead of 8x like other sites have used. Not saying that's good or bad, it's just a new take on things.
      <]q I think so too. There wasn't as much emphasis on the poor quality of Nvidias IQ... I kind of skimmed through it because I'm busy but did TR compare proper modes on these cards? I mean, ATi's performance mode looks better than NVidias Best IQ mode.

      • indeego
      • 18 years ago

      This isn’t Tom’s, and that’s a VERY VERY VERY good thingg{.}g

      • Anonymous
      • 18 years ago

      well the “slightly” quieter version has shifted the noise level from the original 54db to 47db, which is still VERY loud by my standards

    • yokem55
    • 18 years ago

    Umm….Scott, did you mix up the pic’s comparing the image quality of the 43.00 and 43.45’s because the one labeled as being from the 43.00’s look a lot worse to me than the one from the 43.45’s with some nasty artifacts underneath the butterfly. Are you saying nVidia /[

      • Anonymous
      • 18 years ago

      Yeah it improved, but if you read carefully, you’ll notice that Damage said that the new drivers created other problems

    • wesley96
    • 18 years ago

    Hmm, to think that they would have the ‘9700Pro killer’ so much later than the actual volume shipping of 9700Pro occurred is quite damning, and few would disagree.

    Also, this review confims that my money was well spent buying the 9700Pro instead of waiting for 5800Ultra to hit the shelves. The only disappointment(?) I have is that, because NVIDIA’s counterpunch is so weak, the prices on ATI’s high-ends refuse to budge much at all. I bought mine on late February and so far the price has fallen by, ooh, 80 cents here. At least during February I saw the prices fall by around 8 dollars. Perhaps I made the purchase at an appropriate time.

    • Zenith
    • 18 years ago

    /me cries

    And i had such hope for it…and it continues to let me down.

    • Anonymous
    • 18 years ago

    Will Nvidia finally have the 9700pro killer???? lol. 8 months later. Then they have to worry about a increased clock speed 9800 running on 256dd2 memory. Not to mention around the corner, the R400 which is on a complete different chip-set.

    • SuperSpy
    • 18 years ago

    Very nice review, Damage, I applaud your effort to exploit the cards underlying hardware.

    Hopefully, we will see nVidia hit it with the future NV3x designs, and hopefully they won’t be too stuck-up to send a review sample your way this time.

    • indeego
    • 18 years ago

    “It’s louder than a bright plaid leisure suit, and the white noise repeatedly lulled me to sleep as I was trying complete this review.”

    I just chocked on my dinner reading thatg{

      • indeego
      • 18 years ago

      And then I looked through my dictionary and shook my headg{<.<}g

    • eckslax
    • 18 years ago

    In all seriousness, that was the best of the Geforce FX reviews that I have read. You think you got enough shots of Dawn on the front page? =)

      • Anonymous
      • 18 years ago

      Ass Kisser!

    • Forge
    • 18 years ago

    Would have been nice if any useful number had ever hit retail.

    Oh well, NV swears NV35 will be out soon, and will be the 9700 killer NV30 was going to be.

      • TheCollective
      • 18 years ago

      I’ll believe that when I see it.

    • eckslax
    • 18 years ago

    Well, can’t say I was surprised.

      • andurin
      • 18 years ago

      Yeah, I think most of us knew what was coming with this review.

Pin It on Pinterest

Share This