Bjorn3D.com Bjorn3d.com – Satisfying Your Daily Tech Cravings Since 1996
BFG offers an extra 128MB of memory with its FX5600, but is it worth the extra money?
Introduction
It’s time once again to play the name game with NVIDIA’s video cards! For those playing the home version, today’s contestants are MSI’s Ti4800 SE and BFG’s FX 5600. As you may recall from previous episodes, the Ti4800SE is a Ti4400 with AGP 8X support. The Ti4200 has proven to be a very popular contestant amongst bargain seekers. It’s fast out of the box and overclocks well. Although our review of the Ti4800 didn’t reveal any performance advantage for AGP 8X in today’s crop o’ games, it was still a very good performer at a nice price.
Our new player, the FX 5600, sits about half of the way up the current FX series card lineup. The FX 5200 is the bottom tier, followed by the 5200 Ultra, then by the 5600 (roughly occupying the space previously taken by the Ti4400). By virtue of being an FX and having a higher number in its name, the 5600 should wipe the floor with the Ti4800, right? Not so fast, first we have to show you the specs and drop the annoying game show schtick.
Specifications & Features
- GPU: NVIDIA GeForce FX 5600
- Bus Type AGP
- Memory 256MB DDR
- Core Clock 325MHz
- RAMDAC Dual 400MHz
- API Support DirectX, OpenGL for windows
- Connectors VGA, DVI, VIVO
- 81 million vertices/sec.
- AGP 8X (compatible with 4X AGP 2.0 compliant slots)
- 4 pixels per clock rendering engine
- Up to 16 textures per pass
- Intellisample Technology
- Video Mixing Renderer
- Digital Vibrance Control 3.0
- Vertex Shader 2.0+ and Pixel Shader 2.0+
- Geometry Displacement Mapping
- NVDVD 2.0 multimedia software
- Ulead VideoStudio 6 SE video editing software
- Full installation manual (.PDF) with troubleshooting guide
- GeForce FX demos
- VIVO cable
The unique feature of the card is its 256MB of memory. Will it make a difference in today’s games?
The Bundle & The Card
Compared to the bundle included with MSI’s FX 5600-VTDR128 (11 CDs, DVI-I / VGA Adapter, TV-Out / Video-In 1-to-4 Connecting Cable and Infrared Remote Receiver, a remote control, and a case badge, see our review), the bundle of BFG’s 5600 is a tad light. You get:
- Ulead VideoStudio 6 SE video editing software
- One CD with:
- Drivers
- PDF Installation Manual
- PDF nView User’s Guide
- NDVD 2.0
- GeForce Ti and FX demos
- Quick Install Booklet
- DVI-to-VGA adapter (for dual analog monitor support)
- TV-Out / Video-In 1-to-4 Connecting Cable
- One BFG case sticker, two NVIDIA stickers
It’s by no means a bad bundle, but it doesn’t do much to distinguish the card from its competitors.
The card is small, smaller than the Ti4800, and features a pleasing dark blue PCB. The other cosmetic feature is that wacky “We’re Committed” guy emblazoned on the GPU fan.
Installation
There’s (thankfully) not much to say here. The card went in, it worked, no drama. And that’s a good thing.
Testing
Test System Configuration
- AMD AthlonXP 1800+
- EPoX 8RDA+ Motherboard (nForce2)
- 256MB Corsair XMS PC3200 DDR SDRAM, 256MB Corsair PC2400 DDR SDRAM
- NVIDIA Display Driver 44.03
- NVIDIA nForce Unified Driver Package v2.03
- DirectX 9.0a
The B3D benchmark suite utilizes 3DMark2001SE to test older DirectX capability and compatibility. We’re not quite ready to use 3DMark03 given all the issues between NVIDIA and Futuremark. Unreal Tournament 2003 is one of the more demanding applications at this time, and it is easy to get consistent benchmark results using UT2K3Bench 1.4. Return to Castle Wolfenstein uses the comparatively ancient Quake III engine, but it can still test a lower-end card, and it wouldn’t be right not to include an id product in a video card review. 😉 Besides, it’s also easy to get great results from RTCW using Q3Bench. Finally, the Comanche 4 demo may not be DirectX9, but it can still kick some video booty.
All tests were run at three resolutions, one time without antialiasing or anisotropic filtering and another with 4x antialiasing and 8x anisotropic filtering.
3DMark2001SE Build 330
Setting: Default – NoAA, NoAF, Quality Mode
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 |  8701 |
| GeForce4 Ti4800 SE |  10640 |
|
1280×1024
|
|
| BFG FX5600 | 7012 |
| GeForce4 Ti4800 SE | 8804 |
|
1600×1200
|
|
| BFG FX5600 | 5594 |
| GeForce4 Ti4800 SE | 7170 |
The GeForce4 Ti4800 SE is faster at every resolution when advanced features aren’t used.
3DMark2001SE Build 330
Setting: 4xAA, 8xAF, Quality Mode
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 | 5511 |
| GeForce4 Ti4800 SE | 4627 |
|
1280×1024
|
|
| BFG FX5600 | 3571 |
| GeForce4 Ti4800 SE | 2808 |
|
1600×1200
|
|
| BFG FX5600 | 2347 |
| GeForce4 Ti4800 SE | 1918 |
If antialiasing and anisotropic filtering make a difference in your gaming pleasure, then the FX 5600 demonstrates its value once we enable those features.
Unreal Tournament 2003
Setting: NoAA, NoAF, Quality Mode
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 – Bots | 46.04 |
| GeForce4 Ti4800 SE – Bots | 51.56 |
|
1280×960
|
|
| BFG FX5600 – Bots | 31.47 |
| GeForce4 Ti4800 SE – Bots | 42.61 |
|
1600×1200
|
|
| BFG FX5600 – Bots | 25.01 |
| GeForce4 Ti4800 SE – Bots | 35.62 |
The scores are close, but the Ti4800 SE still beats out the FX5600.
Unreal Tournament 2003
Setting: 4xAA, 8xAF, Quality Mode
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 – Bots | 28.00 |
| GeForce4 Ti4800 SE – Bots | 19.68 |
|
1280×1024
|
|
| BFG FX5600 – Bots | 16.06 |
| GeForce4 Ti4800 SE – Bots | 11.30 |
|
1600×1200
|
|
| BFG FX5600 – Bots | 11.18 |
| GeForce4 Ti4800 SE – Bots | 8.10 |
It was certainly interesting watching the slow motion demo at 1600×1200, but I wouldn’t try that mode in a real game. 🙂 If you are CPU-bound like my current configuration, you’d best switch to 800×600 if you need antialiasing or anisotropic filtering.
Return To Castle Wolfenstein – version 1.44, “Checkpoint” demo
Setting: NoAA, NoAF, Quality Mode, Max Configuration
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 | 79.2 |
| GeForce4 Ti4800 SE | 88.4 |
|
1280×1024
|
|
| BFG FX5600 | 61.6 |
| GeForce4 Ti4800 SE | 86.7 |
|
1600×1200
|
|
| BFG FX5600 | 44.0 |
| GeForce4 Ti4800 SE | 80.2 |
Check out those framerates! It is disappointing to see the dramatic difference between the FX 5600 and the Ti4800 at 1600×1200.
Return To Castle Wolfenstein – version 1.44, “Checkpoint” demo
Setting: 4xAA, 8xAF, Quality Mode
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 | 58.9 |
| GeForce4 Ti4800 SE | 61.1 |
|
1280×1024
|
|
| BFG FX5600 | 38.6 |
| GeForce4 Ti4800 SE | 38.5 |
|
1600×1200
|
|
| BFG FX5600 | 26.2 |
| GeForce4 Ti4800 SE | 26.2 |
The Ti4800 SE appears to keep up here, but that’s because it drops down to 2x antialiasing. The FX 5600 does fairly well at 1024×768, but at higher resolutions it’s producing slide shows.
Comanche 4 Demo
Setting: NoAA, NoAF, Quality Mode, Shaders Enabled
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 | 32.58 |
| GeForce4 Ti4800 SE | 33.45 |
|
1280×1024
|
|
| BFG FX5600 | 30.49 |
| GeForce4 Ti4800 SE | 34.64 |
|
1600×1200
|
|
| BFG FX5600 | 25.77 |
| GeForce4 Ti4800 SE | 33.30 |
This demo is a real bit buster and neither card was able to distinguish itself.
Comanche 4 Demo
Setting: 4xAA, 8xAF, Quality Mode, Shaders Enabled
| GFX Card/GPU | Score |
|
1024×768
|
|
| BFG FX5600 | 22.67 |
| GeForce4 Ti4800 SE | 22.88 |
|
1280×1024
|
|
| BFG FX5600 | 16.16 |
| GeForce4 Ti4800 SE | 14.85 |
|
1600×1200
|
|
| BFG FX5600 | 11.48 |
| GeForce4 Ti4800 SE | 10.17 |
You can have antialiasing and anisotropic filtering, but you won’t want to actually play with them on. 🙁
Conclusions
The BFG FX 5600 presents a number of difficulties in attempting to determine a score for the card. We try to judge a product based on the board maker’s implementation of a spec design and the actual design of the chip/chipset. Using those two criteria, the BFG 5600 is only moderately successful.
BFG’s product is stable, attractive, and competitively priced. But it also must be judged against its peers. We can directly compare it to MSI’s implementation of the same NVIDIA design. The scores between cards, when system differences are accounted for, are nearly identical. But, the BFG card has 256MB of memory! Our tests show that the extra memory just isn’t needed for today’s gaming applications and/or that the FX 5600 has enough architectural limitations that prevent it from making full use of the extra memory. So for roughly the same amount of money, MSI offers similar performance and a superior bundle.
The FX 5600 is also problematic. In one sense, it is moderately priced, and it offers DirectX9 features. But will you want this card for your DX9 games such as Doom III? Frankly, I doubt it. The FX 5600 was not able to best the previous generation in raw horsepower. It was more successful than its Ti brethren when antialiasing and anisotropic filtering were applied, but it didn’t set any speed marks in the process. All this leads us to believe that while the FX 5600 will be able to utilize the DX9 features of future games, it may not perform them well enough to satisfy the gamer expecting big results from his FX-class card. At this point of NVIDIA’s FX product offerings, the FX 5200 Ultra represents a better value for performance. In addition, ATi’s mid-range DX9 cards must also be considered; they offer similar or greater performance at a similar price point.
It is with some reluctance that we award the BFG FX 5600 a 7 out of 10, but without the Bjorn3D Seal of Approval. It’s not a bad card by any means, but its performance and bundle do not distinguish it from other NVIDIA card makers, other NVIDIA product lines, and competitor’s products.
Final Score
| Performance |
7
|
This mainstream card outperforms its Ti equivalent with 4XAA and 8XAF settings enabled, but can’t keep up when they aren’t in use. Its performance at higher resolutions is also suspect; high res gamers may want to look at a higher model. |
| Features |
6
|
The card has video-in, video-out and dual monitor capability. It supports DX9 as well. However, the feature you’re paying most for, the extra 128MB of memory, doesn’t have any real benefit in today’s games. |
| Bundle |
6
|
It’s not bad, but it pales in comparison to MSI’s. |
| Quality |
10
|
The card has performed without incident. The drivers have not shown any errors or performance problems. |
| Price |
6
|
A quick price check shows the BFG 5600 listing for approximately $50 more than MSI’s 5600. The extra 128MB of memory must account for that difference. However, our tests didn’t show any current performance advantage for the extra memory. |
| Total: 7.0 | The BFG FX 5600 is an adequate FX-class card, but its comparatively high price and dubious top feature (extra memory) make us hesitant to recommend it. | |
