iXBT Labs :: Computer Hardware In Detail

Platform

Video

Multimedia

Mobile

Other

CommentsRegister an account
Login:      
Password: Lost password?

XGI Volari Duo V8 Ultra 256MB Video Card Review


 Forum


 Comment





[ Part 1 ]

 

CONTENTS

  1. Test results: Quake3 ARENA
  2. Test results: Serious Sam: The Second Encounter
  3. Test results: Return to Castle Wolfenstein
  4. Test results: Code Creatures DEMO
  5. Test results: Unreal Tournament 2003
  6. Test results: AquaMark3
  7. Test results: RightMark 3D
  8. Test results: Tomb Raider: Angel of Darkness
  9. Test results: Half-Life2 (Beta)
  10. Test results: HALO
  11. Test results: Unreal II
  12. Summary on the test results
  13. 3D quality: anisotropic, bilinear and trilinear filtering
  14. 3D quality in general
  15. Conclusion

 

3D graphics, game tests

For estimation of the performance in 3D games we used: 

  • Return to Castle Wolfenstein (MultiPlayer) (id Software/Activision) - OpenGL, multitexturing, Checkpoint-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, Grand Cathedral demo, test settings: quality, S3TC OFF 

  •  
  • Quake3 Arena v.1.17 (id Software/Activision) - OpenGL, multitexturing, Quaver, 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 2003 v.2225 (Digital Extreme/Epic Games) - Direct3D, Vertex Shaders, Hardware T&L, Dot3, cube texturing, default quality 

  •  
  • Code Creatures Benchmark Pro (CodeCult) test demonstrates operation of cards in the 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 the game scenes) - DirectX 8.1, Dot3, cube texturing, shadow buffers, vertex and pixel shaders (1.1, 1.4). 

  • Tomb Raider: Angel of Darkness v.49 (Core Design/Eldos Software) - DirectX 9.0, three recorded demo. Highest quality, Depth of Fields PS20 off. Test settings are here.

  •  
  • HALO: Combat Evolved (Microsoft) - Direct3D, Vertex/Pixel Shaders 1.1/2.0, Hardware T&L, highest quality

  •  
  • Half-Life2 (Valve/Sierra) - DirectX 9.0, two different recorded demo (ixbt07 and coast). Tests carried out with anisotropy enabled, and with AA and anisotropy enabled.

  •  

If you need the demo benchmarks please email me. 

Quake3 Arena, Quaver

 








Serious Sam: The Second Encounter, Grand Cathedral








Return to Castle Wolfenstein (Multiplayer), Checkpoint

 

 








Code Creatures

 








Unreal Tournament 2003

 








AquaMark3

 




RightMark 3D

 








TR:AoD, paris5_4

 








Half-Life2 (beta), ixbt0703

 

 
 






HALO

 

 






Unreal II

 

 








Summary on the test results

Obviously, XGI's card looks very poor! It falls behind even middle-end FX 5700 Ultra and RADEON 9600 XT, especially in shader applications.

Even without renaming EXE files of the games it's clear that there are optimizations and cheats for certain applications, since in Splinter Cell Volari got a second wind and even wins somewhere. However, the rendering quality... but we will speak a bit later about it. 

3D graphics quality

Before we estimate general rendering quality let's see how anisotropy, bilinear and trilinear filtering affect it. 

ANISOTROPIC, TRILINEAR and BILINEAR FILTERING

Quality is estimated in the RightMark3D, in Pixel Filling test.

Remember that Volari supports only 4x degree of the anisotropic filtering.
 

GeForce FX 5900 Ultra RADEON 9800 PRO Volari Duo V8 Ultra
Bilinear









Trilinear










 

As you can see, Volari demonstrates awful bilinear filtering, while the trilinear one is ok.
 

ANISOTROPIC and angle of inclination GeForce FX 5900 Ultra RADEON 9800 PRO Volari Duo V8 Ultra
ANISOTROPIC without Trilinear
ANISO 8xQ/16xQ/4x, 0 degrees








ANISO 8xQ/16xQ/4x, 30 degrees








ANISO 8xQ/16xQ/4x, 45 degrees








ANISO 8xQ/16xQ/4x, 60 degrees








ANISOTROPIC with Trilinear
ANISO 8xQ/16xQ/4x, 0 degrees








ANISO 8xQ/16xQ/4x, 30 degrees








ANISO 8xQ/16xQ/4x, 45 degrees








ANISO 8xQ/16xQ/4x, 60 degrees








 

3D graphics quality in general

Originally the test tools included the Far Cry demo version which revealed the beauty of shader technologies.

Below you will see why we removed it from the performance tests.

GeForce FX 5900 Ultra RADEON 9800 PRO Volari Duo V8 Ultra
3DMark03, Game1









3DMark03, Game2









3DMark03, Game3









3DMark03, Game4









AquaMark3









Half-Life2










 

Volari's rendering quality is not that good. Does it make sense to use cheats and look for ways to change quality when the performance level is low anyway? - I think not :-(.
 

GeForce FX 5900 Ultra RADEON 9800 PRO Volari Duo V8 Ultra
Far Cry, Example 1, No ANIS, No AA









Far Cry, Example 2, No ANIS, No AA






-
Far Cry, Example 3, No ANIS, No AA






-
Far Cry, Example 4, No ANIS, No AA






-
Far Cry, Example 5, No ANIS, No AA






-
Far Cry, Example 6, No ANIS, No AA






-
Only this card considers that events in the game take place at the moonless night. Well, the demo version was released a short time ago, and they didn't have enough time to bring in new cheats. Plus, it's clear that they do not collaborate with the game developers. 

Conclusion

The Club3D Volari Duo V8 Ultra is still a raw product. How could they release such a card? The drivers are unsteady and need a lot of work, and the architecture itself does not realize its capabilities. Although the fillrate shows that both chips work it looks as if we had a uniprocessor card in the real applications. Even one chip with 8 pipelines and clocked at 330 MHz must look much better! Maybe the second chip is spare? If the first one burns the second could replace it, is it so? But what for does the card have all those internal buses? :)

The card's beautiful appearance hides so many bugs that it's impossible for the drivers to correct them. Probably, it's the chip developers or driver makers that should be blamed. Even 50% of the card's potential is not realized. We hope that new XGI's products will be more completed. 

  1. This products doesn't cost even $150 from the performance standpoint;
  2. This products doesn't cost even $50 regarding compatibility and stability of operation. 
  3. No prospects for performance growth as the main problems is hidden in the accelerator's architecture.
  4. Probably, XGI got income only from selling samples of such cards to reviewers because they hardly shared their samples but they are always ready to sell them ;-)
  5. Judging by the number of companies (1-2 at most) that went with this product Volari will be "very popular".
  6. I feel very sorry for Club3D which got into a mess with this product...

We can recommend this card only for hunters for hardware rarities and curious samples. The company's slogan seems to be like this: "why to pay three times less for the speed if there's such a lame dual-chip card".

After the article went live we received information about numerous video cards makers' intention to introduce dual-chip Volari Duo V8 Ultra products priced below $200. In case it happens, there will be rather illusive hope for XGI to gain popularity among users.

Moreover, on the 11th of February a new Reactor driver version (1.02.05) was released. The fresh drivers correct some bugs in certain games and benchmarks. Unfortunately, issues with FarCry were not resolved at this time.  




 
Andrey Vorobiev (anvakams@ixbt.com)
  Alexander Medvedev (unclesam@ixbt.com


Comments (0)  

  Most Popular Reviews More   RSS  
  Latest Reviews More   RSS  
  Latest News More   RSS  
  Sponsored Links Get listed  
 

Platform  ·  Video  ·  Multimedia  ·  Mobile  ·  Other  ·  Compare Prices  ||  Feedback  ·  Advertise at iXBT Labs  ·  About us  ·  Affiliates  ·  Forum


23

Copyright © Byrds Research & Publishing, Ltd., 1997—2008. All rights reserved.