iXBT Labs - Computer Hardware in Detail

Platform

Video

Multimedia

Mobile

Other

Gainward FX Powerpack Ultra/1000 Golden Sample
(NVIDIA GeForce FX 5800 Ultra) &
Gainward FX Powerpack Pro/660 TV/DVI
(NVIDIA GeForce FX 5200) Video Cards Review






CONTENTS

  1. General information
  2. Video card features
  3. Test system configuration
  4. Test results
  5. Conclusion

As we keep on studying new solutions from NVIDIA, I must say that it is not latecomer NV30 (GeForce FX 5800) which comes to the forefront this spring. Today the attention is focused on the new line supporting DX9 which is positioned for budget and mainstream sectors. These video cards will be the most interesting for the majority. 

ATI launched similar accelerators for the mainstream market (RADEON 9500/PRO) half a year ago. But the sub-$100 sector is burdened with old solutions. The only processor that ATI has there is RADEON 9000 which brought DX81 into the budget sector. But as you are well aware, even the most favorable market conditions do not stimulate the Canadian company to carry out the aggressive policy. It always stumbles over something: bad drivers, shortage of chips, shortage of common sense... That is why GeForce4 MX, though a weaker runner, is still much more popular with users than the advanced RADEON 9000. 

But time goes on, and even non-advanced users realize how weak GF4MX is in supporting shaders and other modern functions, and the year of 2003 is the high time to do away with this anachronism. That is why the market is now filling with inexpensive GeForce FX 5200 based video cards. One of such cards will be tested today. GeForce FX was already studied in our previous reviews: 

That is why we won't dwell on technical parameters and quality issues of 3D functions. 

Today I'm going to look at two aspects: for GeForce FX 5800 Ultra (Gainward's card) - video card performance vs. CPU clock speed; for GeForce FX 5200 - comparison of performance of this card with RADEON 9000/PRO and GeForce4 MX 440-8x. 

Gainward is an established brand on the market, especially its Golden Sample series which includes samples which run reliably at a little overclocked speed or which are bundled with peculiar accessories. Briefly, the company is headquartered in Taiwan for over 20 years, but earlier it dealt with OEM orders only. As an independent manufacturer and seller of its products under PowerPack trademark (earlier CARDexpert) it rang out quite recently. 

Today we are testing two cards: the top solution in the GeForce FX line - 5800 Ultra, and the lowest one - 5200. The former is meant for $350+ niche while the latter is positioned for the sub-$100 market sector. Both accelerators differ only in the speed characteristics (GeForce FX 5200 has only one restriction in AA - 4xS, in contrast to 8xS of 5800). 

But they are far different in the appearance and design. 

Cards

Gainward FX Powerpack Ultra/1000 Golden Sample 






Gainward FX Powerpack Pro/660 TV/DVI 







 
 
 
Gainward FX Powerpack Ultra/1000 Golden Sample 
The card has AGP x2/x4/x8 interface, 128 MB DDR SDRAM (8 chips on both PCB sides). 128bit memory exchange interface. 

Samsung K4N26323AE-GC1K, memory chips of BGA form-factor. Maximum clock speed is 550 (1100) MHz, hence 1.8 ns access time. The memory works at 500 (1000) MHz in 3D and at 300 (600) MHz in 2D.




Gainward FX Powerpack Pro/660 TV/DVI 
The card has AGP x2/x4/x8 interface, 128 MB 128bit memory chips. 

Samsung memory chips of TSOP form-factor. 5ns access time. It corresponds to 200 (400) MHz, and the memory works at this clock speed. The core is clocked at 250 MHz. 





 
 
 
Comparison with the reference design, front view 
Gainward FX Powerpack Ultra/1000 Golden Sample  Reference card NVIDIA GeForce FX 5800 Ultra 






Gainward FX Powerpack Pro/660 TV/DVI  Reference card NVIDIA GeForce FX 5200 Ultra 










 
 
 
Comparison with the reference design, back view
Gainward FX Powerpack Ultra/1000 Golden Sample  Reference card NVIDIA GeForce FX 5800 Ultra 






Gainward FX Powerpack Pro/660 TV/DVI  Reference card NVIDIA GeForce FX 5200 Ultra 







 

Gainward's GeForce FX 5800 Ultra is a copy of the reference card, but for the cooler. GeForce FX 5200 differs much from the reference 5200 Ultra based card. This is not the Ultra version, and the PCB looks much plainer. At the same time, the design of 5200 follows the reference one in every detail: 




As you can see, GF FX 5200 comes with the outdated 5ns memory. If you remember, even GeForce4 MX had 4ns in most cases (I don't mean numerous noname MX440SE cards with 6ns and 7ns memory). This is obviously the cost of 128 MB, and I'm sure that 64MB GF FX 5200 cards will have 4ns memory. 

Therefore, overclocking of such memory (5ns, Samsung) is hardly possible. 

Now have a look at the coolers. 

Gainward FX Powerpack Ultra/1000 Golden Sample 
This cooler is a zest of this card. It looks like NVIDIA's FlowFX, but here air is sucked in from inside the PC case, not from the outside, and an air flow doesn't make such a sudden turn in the plastic tube like in FlowFX, and, therefore, makes less noise. 

But this cooler has its own downsides: 

  • Although the cooler is less noisy, the noise frequency is higher because of the speedy fan, and you can hear it whistling (owners of computers with SCSI hard drives of 10000 and over rpm are well aware of it :-). 
  • Air is taken from inside the PC case which makes cooling less effective. 

Besides, the cooler still switches on/off depending on the graphics mode (2D/3D) and GPU's temperature, though it would be better if a user could decide himself whether to allow such reswitching or not. Such changes irritate much. 













Gainward FX Powerpack Pro/660 TV/DVI 
This card has only a passive pin heatsink, which is sufficient for such frequencies (250 MHz). 



 
 
 
Under the heatsink we can find the processor itself: 



I tried the TV-out of GF FX 5200 and found it pretty good. Aleksei Samsonov and Dmitry Dorofeev, our experts in this sphere, will soon provide more detailed information on this aspect in the updated material on TV-out of GeForce2/3/4 based cards. 

Accessory packs: 
 

Gainward FX Powerpack Ultra/1000 Golden Sample 
This bundle is luxurious. Here you can find a user guide, CD with drivers and utilities, CDs with outdated (unfortunately) games and WinDVD 4.0, S-Video-to-RCA and DVI-to-d-Sub adapters, and TV-out extenders. 

The most peculiar thing is a 5.1-channel sound card (but I think that those who can afford a card of $400-450 won't need such cheap sound; they'd better go with something like Creative Audigy(2)) 

Also, there is a FireWire controller, which comes in handy here, because users buying such expensive accelerators usually have digital video cameras. 






















Gainward FX Powerpack Pro/660 TV/DVI 
Here we have a user guide, CD with drivers and utilities, WinDVD 4.0, S-Video-to-RCA adapter, TV-out extenders, and DVI-to-d-Sub adapter. 



 
 
 
Both cards ship in retail packages. 
Gainward FX Powerpack Ultra/1000 Golden Sample 
The designers at Gainward have finally invented something new. This time they put on the box a pretty, though rendered, girl. 


Gainward FX Powerpack Pro/660 TV/DVI 
But I think it's wrong to use the same box for so different cards, in spite of the same line. One sticker with the card's name is not enough. 



 
 
 

Overclocking

Gainward FX Powerpack Ultra/1000 Golden Sample  The overclocking is not fruitful - after 540 MHz the chip hangs, while the memory's summit is only 1040 MHz. 
Gainward FX Powerpack Pro/660 TV/DVI  While the chip goes up noticeably - up to 315 MHz, the memory has clambered up to 405 MHz from 400 MHz. Later you will see that the card's performance is much limited by the memory, and GPU alone does not boost the performance. 

 
 
 

Note that 

  • Overclocking requires additional card cooling (for its memory, in particular): 



  • Overclocking depends on a certain sample, and you shouldn't extend single-card results to the entire series or trade mark. Overclocking results are not obligatory characteristics of video cards. 

Testbed and drivers

Testbed: 

  • Pentium 4 3066 MHz based computer: 
    • Intel Pentium 4 3066 MHz; 
    • ASUS P4G8X (iE7205) mainboard; 
    • 1024 MB DDR SDRAM; 
    • Seagate Barracuda IV 40GB hard drive; 
    • Windows XP SP1; DirectX 9.0; 
    • ViewSonic P810 (21") and ViewSonic P817 (21") monitors. 
    • NVIDIA's drivers v43.45. 

  • Athlon XP 2600+ based computer: 
    • AMD Athlon XP 2600+ (2133 MHz); 
    • EPoX (nForce2) mainboard; 
    • 1024 MB DDR SDRAM PC3200; 
    • Seagate Barracuda IV 40GB hard drive; 
    • Windows XP SP1; DirectX 9.0. 
    • ViewSonic P810 (21") and ViewSonic P817 (21") monitors. 
    • NVIDIA's drivers v43.45. 

VSync off in drivers, texture compression off in applications. Texture detail set to High Quality. 

Video cards for comparison: 

  • Gainward Powerpack Pro/650 (GeForce4 MX440-8x, 275/256 (512) MHz, 64 MB); 
  • Hercules 3D Prophet 9000 PRO (RADEON 9000 PRO, 275/275 (550) MHz, 128 MB, driver 6.307); 
  • Sapphire Atlantis RADEON 9000 (RADEON 9000, 250/200 (400) MHz, 128 MB, driver 6.307) 
  • Hercules 3D Prophet 9700 PRO (RADEON 9700 PRO, 325/310 (620) MHz, 128 MB, driver 6.307); 
  • ATI RADEON 9800 PRO (380/340 (680) MHz, 128 MB, driver 6.307). 

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. 

As for the sample tested, together with the ViewSonic P817 monitor and BNC Bargo cable it showed excellent quality at the following resolutions and clock speeds:

Gainward FX Powerpack Ultra/1000 Golden Sample  1600x1200x85Hz, 1280x1024x120Hz, 1024x768x160Hz 
Gainward FX Powerpack Pro/660 TV/DVI  1600x1200x85Hz, 1280x1024x120Hz, 1024x768x160Hz 

 

3D performance

Test applications: 

     
  • 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 

  •  
  • Return to Castle Wolfenstein (MultiPlayer) (id Software/Activision) - OpenGL, multitexturing, Checkpoint-demo, maximum test settings, S3TC OFF, the configs are available here
    •  

  • Serious Sam: The Second Encounter v.1.05 (Croteam/GodGames) - OpenGL, multitexturing, Grand Cathedral demo, test settings: quality, S3TC OFF 
    •  

  • Unreal Tournament 2003 Demo v.1077 (Final Release) (Digital Extreme/Epic Games) - Direct3D, Vertex Shaders, Hardware T&L, Dot3, cube texturing, default quality 

  •  
  • 3DMark2001 Pro (MadOnion/Remedy) - DirectX 8.0, Hardware TCL, Game1, Game2, Game3, Game4; Low, High detail levels

  •  
  • RightMark 3D (one of the game scenes) - DirectX 8.1, Dot3, cube texturing, shadow buffers, vertex and pixel shaders (1.1, 1.4). 

Today we have two aspects for analysis: 

  1. Performance of GeForce FX 5800 Ultra vs. CPU's frequency (AthlonXP); 
  2. Comparison of performance of GeForce FX 5200 without AA and anisotropy enabled. 

3D graphics, 3DMark2001 game tests

 

 
 

3DMark2001, 3DMARKS

















3DMark2001, Game1 Low details

















3DMark2001, Game2 Low details

















3DMark2001, Game3 Low details

















3DMark2001, Game4
















I guess, it's clear how NV30's performance depends on the CPU's clock speed, and you can draw the conclusion yourselves. 

In most cases GeForce FX 5200 lags behind GeForce4 MX440-8x running between RADEON 9000 and 9000 PRO. It's quite fair for the configuration with two 2 texture units and the frequency of 250 MHz in contrast to 4 texture units and 275 MHz of MX440-8x. GeForce FX 5200 works at the same frequency as RADEON 9000 but it performs better. Besides, being a flexible system, it depends much on optimization of the drivers. That is why the speed can grow even more. 

3D graphics, game tests

 

 
 

Quake3 Arena

















Serious Sam: TSE

















RtCW

















UT2003 DEMO

















Rightmark 3D v.0.4
















Sometimes GeForce FX 5200 loses even to RADEON 9000, but sometimes its performance is really superb. I think NVIDIA will keep on polishing the drivers, and the vendors should think about cutting prices for such cards to create the demand and to make users forget about such an anachronism as GeForce4 MX. 

In closing I must say that both cards worked perfectly, and had no problems with stability. 

Conclusion

Such cards as GeForce FX 5200 can be good substitutes for GeForce4 MX, and they successfully fight against RADEON 9000. The manufacturers had better replace 128 MB memory with 64MB and reduce the sale prices. The sellers should realize that this is not GeForce FX 5800, but only its shadow bundled with all the same functions. This accelerator can be competitive only for GeForce4 MX and RADEON 9000. This fact should be accounted for in the price. 

On the one hand, GeForce FX 5800 Ultra is a solid card with a striking accessory pack. But some accessories are not needed for users, and the cooler produces irritating high-pitch noise, though it actually works quieter. Let's wait and see what the prices are going to be. I must say that NVIDIA keeps on optimizing its drivers, and NV30 has already raised its speed. But there are still problems with AA: certain AA modes do not work in some games or they are replaced with lower ones. 

Highs:

  • Almost the best 3D performance among the today's accelerators which are already available for buyers (I mean GeForce FX 5800 Ultra); 
  • GeForce FX 5200 has performance comparable to GeForce4 MX 440 and RADEON 9000 (also remember that it supports all kinds of shaders); 
  • High build quality; 
  • Stable and reliable operation; 
  • A rich accessory pack of Gainward's GeForce FX 5800 Ultra (though some of the accessories are not needed) 

Lows:

  • Irritating noise of the cooler of GeForce FX 5800 Ultra; 
  • GeForce FX 5800 Ultra is in short supply and overpriced; 
  • Overpriced GeForce FX 5200 (for the beginning); 
  • Unnecessary 128 MB on FX 5200 (64MB and faster memory would be better; at least, overclockers would like it :-) 


 

Andrey Vorobiev (anvakams@ixbt.com
 

Write a comment below. No registration needed!


Article navigation:



blog comments powered by Disqus

  Most Popular Reviews More    RSS  

AMD Phenom II X4 955, Phenom II X4 960T, Phenom II X6 1075T, and Intel Pentium G2120, Core i3-3220, Core i5-3330 Processors

Comparing old, cheap solutions from AMD with new, budget offerings from Intel.
February 1, 2013 · Processor Roundups

Inno3D GeForce GTX 670 iChill, Inno3D GeForce GTX 660 Ti Graphics Cards

A couple of mid-range adapters with original cooling systems.
January 30, 2013 · Video cards: NVIDIA GPUs

Creative Sound Blaster X-Fi Surround 5.1

An external X-Fi solution in tests.
September 9, 2008 · Sound Cards

AMD FX-8350 Processor

The first worthwhile Piledriver CPU.
September 11, 2012 · Processors: AMD

Consumed Power, Energy Consumption: Ivy Bridge vs. Sandy Bridge

Trying out the new method.
September 18, 2012 · Processors: Intel
  Latest Reviews More    RSS  

i3DSpeed, September 2013

Retested all graphics cards with the new drivers.
Oct 18, 2013 · 3Digests

i3DSpeed, August 2013

Added new benchmarks: BioShock Infinite and Metro: Last Light.
Sep 06, 2013 · 3Digests

i3DSpeed, July 2013

Added the test results of NVIDIA GeForce GTX 760 and AMD Radeon HD 7730.
Aug 05, 2013 · 3Digests

Gainward GeForce GTX 650 Ti BOOST 2GB Golden Sample Graphics Card

An excellent hybrid of GeForce GTX 650 Ti and GeForce GTX 660.
Jun 24, 2013 · Video cards: NVIDIA GPUs

i3DSpeed, May 2013

Added the test results of NVIDIA GeForce GTX 770/780.
Jun 03, 2013 · 3Digests
  Latest News More    RSS  

Platform  ·  Video  ·  Multimedia  ·  Mobile  ·  Other  ||  About us & Privacy policy  ·  Twitter  ·  Facebook


Copyright © Byrds Research & Publishing, Ltd., 1997–2011. All rights reserved.