AGP x8/x4/x2 interface, 128 MB DDR SDRAM in 8 chips on the front PCB side.

Here are some aspects to be pointed out:

1. The card is a copy of the reference sample

2. The memory size was cut down from 256MB to 128MB - on the backside there are empty space for missing BGA chips. But the 256bit bus wasn't shortened because the card has 8 32bit memory chips (256 bits in all). The card just doesn't support the dual-bank memory mode.
   

Now let's have a look at the cooler.

Albatron Gigi GeForce FX 5900PV 128MB
The cooler consists of a copper-based massive heatsink equipped with three fans with impellers covered with dye shining in the ultraviolet rays. The cooler has a LED in the center which can be seen through the drop-like hole in the center of the heatsink cover. This most interesting feature of this cooler is a three-fan system, with two fans working simultaneously and the third fan being a spare one. On the photo it's on the right. The fan controller turns on all three fans when the PC starts up, but after taking the measurements of the first two fans, it disables the third one. The fan controller (a logic device) is situated in the heatsink's center on a small PCB. The cooler is quiet, that is why the advertised noise reduction system is not fiction. Besides, the heatsink's cover looks like a flying bird.

The card is equipped with the external Philips 7108 codec to control the Video-In. This model supports the VIVO indeed. But the TV-out is realized in the NV35.

Have a look at the NV35:

It was made on the 20th weak of this year (end of May) after the announcement of the GeForce FX 5900, that is why the codename is replaced with the FX 5900.

The cards of this type are not small

The cooler, though shrinked in all dimensions, can be an obstacle for installation of a PCI card into the first slot:

If a PCI card has peripheral components underside it can't be inserted into the first slot. Besides, such card wouldn't allow for proper cooling of the video card.

Now let's change a little our order of card examination and look at the package and then at what's in the box.

The card comes in a retail package, just look at it!

If you take off the cardboard cover from the bag, you will see the embroidered logo and the company's name.

Now the accessories:

Testbed and drivers

Testbed:

• Pentium 4 3200 MHz based computer:
• Intel Pentium 4 3200 MHz CPU;
• DFI LANParty Pro875 (i875P) mainboard;
• 1024 MB DDR SDRAM;
• Seagate Barracuda IV 40GB HDD;
• Windows XP SP1; DirectX 9.0a;
• ViewSonic P810 (21") and ViewSonic P817 (21") monitors.
• NVIDIA drivers v44.65.

VSync off, S3TC off in applications.

Cards used for comparison:

• Hercules 3D Prophet 9800 PRO (RADEON 9800 PRO, 380/340 (680) MHz, 128 MB, driver 6.360).

Overclocking

Note that:

• during the overclocking you should provide additional cooling, in particular, for the card (first of all, for its memory):

overclocking depends on a sample, and you shouldn't generalize the results of one card to all video cards of this trade mark or series. The overclocking results are not the obligatory characteristics of video cards.

Test results

Before we start examining 2D quality, I should say there are no complete techniques for objective 2D quality estimation because:

1. 2D quality much depends on certain samples for almost all modern 3D accelerators;
2. Besides videocards, 2D quality depends on monitors and cables;
3. Moreover, certain monitors might not work properly with certain video cards.

With the ViewSonic P817 monitor and BNC Bargo cable the card showed excellent quality at the following resolutions and clock speeds:

In general, the 2D quality is excellent (the quality of this sample in our conditions).

Test results: cards' performance

Conventional signs: ANISO 8xP - Anisotropic 8x Performance (earlier it was called Balanced), ANISO 8xQ - Anisotropic 8x Quality, ANISO 16xQ - Anisotropic 16x Quality.

Test applications:

• Return to Castle Wolfenstein (MultiPlayer) (id Software/Activision) - OpenGL, multitexturing, ixbt0703-demo, test settings - maximum, S3TC OFF, the configurations can be downloaded from here
• Serious Sam: The Second Encounter v.1.05 (Croteam/GodGames) - OpenGL, multitexturing, ixbt0703 demo, test settings: quality, S3TC OFF
• Quake3 Arena v.1.17 (id Software/Activision) - OpenGL, multitexturing, ixbt0703 demo, test settings - maximum: detailing level - High, texture detailing level - #4, S3TC OFF, smoothness of curves is much increased through variables r_subdivisions "1" and r_lodCurveError "30000" (at default r_lodCurveError is 250 !), the configurations can be downloaded from here
• Unreal Tournament v.2225 (Digital Extreme/Epic Games) - Direct3D, Vertex Shaders, Hardware T&L, Dot3, cube texturing, default quality
  
  
• Code Creatures Benchmark Pro (CodeCult) - the game that demonstrates card's operation in DirectX 8.1, Shaders, HW T&L.
  
  
• Unreal II: The Awakening (Legend Ent./Epic Games) - Direct3D, Vertex Shaders, Hardware T&L, Dot3, cube texturing, default quality
  
• RightMark 3D v.0.4 (one of game scenes) - DirectX 8.1, Dot3, cube texturing, shadow buffers, vertex and pixel shaders (1.1, 1.4).
  

If you want to get the demo benchmarks that we use let me know by e-mail.

Today we also studied the difference between official release of the drivers v44.03 and latest beta version (WHQL) 44.71.

Quake3 Arena

The drivers show no difference. The FX 5900 wins from the RADEON 9800 PRO in the heavy modes at the expense of AA and Performance-level anisotropy (we compare this anisotropy level of the GeForce FX with the Quality level of the RADEON 9800 because in this case both functions have some reductions: a lower anisotropy degree for certain angles in case of ATI, and simpler trilinear filtering and anisotropy on far textures in case of NVIDIA; that is why their pluses and minuses are more or less balanced, though we also compare Quality levels of both cards for those who are sure that ATI's "crippled" anisotropy paired with the normal trilinear filtering should be compared only to NVIDIA's normal anisotropy).

Serious Sam: The Second Encounter

No difference between the drivers again (note that the 3DMark03 demonstrates a noticeable difference. Think again what the 3DMark03 measures indeed). Again, the FX5900 outdoes the RADEON 9800 PRO in the heavy mode with the AA and performance-level anisotropy, but loses to the latter in the light modes.

Return to Castle Wolfenstein (Multiplayer)

If you remember, all GeForce FX cards were much behind their competitors starting from the drivers 43.00 to 44.03, and only with the drivers 44.65 the programmers have cured that. Note that the rise wasn't caused by the cheats in the standard Chechpoint demo because in our case the scores grow up as well.

In the fight against the RADEON 9800 PRO NVIDIA's baby wins again only in the modes with AA and Performance-level anisotropy.

Code Creatures

The difference in the drivers is considerable. Up to the driver v44.71 the anisotropy didn't work in this test. I demonstrated that last time. Starting from the v44.71 this function works correctly, hence such a performance drop. By the way, it shows the approximate performance consumed by this function in this test. It looks verisimilar because there are a lot of polygons and surfaces to be processed (but maybe it's better to optimize it as one would hardly notice anisotropy on the vertically growing grass).

The card beats the RADEON 9800 PRO only in case of AA and a little in case of the AA+Anisotropy Performance.

Unreal Tournament 2003

The new version has lifted up the scores a little.

The RADEON 9800 PRO takes the lead in all modes except the Performance-level anisotropy.

Unreal II: The Awakening

There is no difference between the drivers, and the FX 5900 is far behind its rival in all modes.

RightMark 3D

RightMark, a packet richest in shaders, demonstrates how weak shader units are of the FX 5900. No difference in the drivers.