Monster-Hardware Logo
  Login or Register
Real American

USA Flag

 
nukeNAV
 
Search



 
Who's Online
There are currently, 25 guest(s) and 0 member(s) that are online.

You are an Anonymous user. You can register for free by clicking here
 
Latest Forum Posts
Last 10 Forum Messages

My Best Buy Sales Experience
SABERTOOTH R2.0 BIOS MIA
SodaStream Anyone?
Tazerball
A Joke Gone Bad
Retrode Gaming
Stealth Bastard
Cooler Master HAF X
Thoughts About Bulldozer
Crucial DDR3-1600 8GB

Monster-Hardware Forums

 
Latest Reviews

· Crucial DDR3-1600 8GB
· Cooler Master HAF X
· Seagate ST31000524AS
· Logisys MS6801 Mouse
· Mitsuba DC500T 5MP
· Infrared Thermometer
· DXG-595V 5MP 1080p
· SanDisk Sansa View
· Rfrb Dell Latitude D410
· (HP) Sempron LE-1250
 
NukeSentinel
This is the list of NukeSentinel(tm) banned IP addresses.

· 77.110.147.88  - Filter
· 176.52.117.27  - Filter
· 202.28.119.237  - Filter
· 176.52.90.19  - Filter
· 176.52.110.133  - Filter
· 174.139.163.82  - CLike
· 89.251.152.14  - Filter
· 153.121.47.69  - Script
· 216.228.104.39  - Script
· 212.83.157.218  - Script
· 58.64.150.234  - Script
· 176.52.101.188  - Filter
· 83.222.254.18  - Script
· 94.23.232.18  - Script
· 64.90.57.127  - Script
· 27.159.229.200  - Script
· 49.212.176.177  - Script
· 216.240.36.214  - Script
· 85.90.193.224  - Script
· 27.159.234.122  - Script

NukeSentinel(tm)
 

Palit Geforce FX 5200

Welcome once again from the secret Monster-Hardware labs! Today, we are going to take a look at a mid-range video card – the Palit Daytona GeForce FX 5200 Ultra. This card has been designed to take advantage of the consumer market segment that wants reasonable performance at a value price, in a card that can still be useful at the next level when DirectX 9.0 applications start to get stronger in the software market.

Palit Front     Palit Back

First, I think it is in order to give a bit of background. I am alone among my Monster-Hardware brethren in that I am NOT a huge gaming participant. Wife, kids, the job…all conspire to deny me the fun that is so rightfully mine to have. I’ll play some games occasionally when I get the chance, and also my kids use my computer when they need to play something that overwhelms their PIII-500. Secondly, I am famous for being frugal. EXTREMELY frugal. I’d rather spend a week and $50 in parts making something myself than spend $120 to buy it off the shelf (ask Outcast sometimes for a picture of his industrial strength 18VDC peltier power supply!). I simply love to get the most value out of a piece of equipment as long as it meets my needs – heck, in my early days at Monster-Hardware when everyone else had 1.4G Thunderbirds, I was sputtering along with an AMD K6-2/300 overclocked to 366!

So, with that in mind, I am exactly the type customer had in mind when they created the GeForce FX 5200 line. I could never cost justify buying an ATI Radeon 9700 All in Wonder – but a card in this range is more up my alley. It is a magnitude leap above the ATI 32MB Rage 128 card I had installed. The GeForce 5200 Ultra does provide some nice features for it’s price range:

GeForce FX 5200 Ultra

-Chip: NV34 (150 NM process technology)
-Memory: 128 MB DDR
-Bandwidth: 10.4 GB/sec
-AGP: 8x
-DirectX: DirectX 9.0 support
-Outputs: Standard SVGA 15 pin, S-Video, DVI
-Miscellaneous: Integrated TV and MPEG-2 encoders
-Core clock - 325 Mhz
-Memory speed - (2 x 325) 650 Mhz

These are significant core clock and memory speed improvements over the GeForce FX 5200 non-Ultra version, which sports a core clock of 250 Mhz and memory speed of 400 Mhz. As we might expect, performance is sure to be improved. Additional changes from higher end models include the fact that there are 4 pixel pipelines instead of 8 for the 5800 series, and the memory for the 5200 series is typically DDR instead of DDR-II.

Palit Outputs

The contents of the video card package is a little slim, but that’s really all I expected for a “value” card. The box included the card itself, an S-Video cable about 3 feet long, and two compact discs. The first is the driver disc, which includes drivers for Windows 9x, ME, NT, 2000, and XP. The second disc includes a bundled version copy of Cyberlink’s PowerDVD XP 4.0 software, pretty standard fare for a new video card.

Palit Stuff

When I took a look at the contents of the included driver disc, I was a bit disappointed. The user manual that was included runs to about 60 pages – but essentially, it is a 4-1/2 page manual that is repeated in 11 different languages. The minimal contents include sections on Minimum requirements, Driver/DirectX installation, and a very minimal Troubleshooting section. Obviously, installing a video card for most people is simply not that big a deal – but for a newbie, or someone that experienced any sort of problem, this might be a little worrisome. Copies of DirectX 8.1 and 9.0 are both included on the disc. The included drivers were Nvidia version 43.51, so I jumped over and downloaded the 44.03 drivers and had my files ready to go when I put the card in. Additionally, I went ahead and downloaded the Coolbits hack for making memory adjustments, ifyaknowwhutImean. Next, it was time for some testing. To begin, let me give the base system specs for the parts that matter:

Epox 8K9A2+
AMD 1700+ overclocked to 2000+
512MB generic DDR RAM
(2) Maxtor 40 GB hard drives 5400 RPM
Windows XP operating system
DirectX 9.0
Generic 17” CRT monitor

As I mentioned, I don’t exactly have the latest and greatest, but this system is probably pretty representative of the target market. And before I get any flak about the 5400 drives, I got them on clearance - $9.50 each, new in the box. That’s a GREAT performance to pocketbook ratio!

Once again, with an eye towards the target market, we’re going to run some simple tests with Unreal Tournament 2003 Demo, Futuremark’s 3Dmark2001SE, and 3Dmark2003 build 320. While the 3Dmark2003 might not be the best comparison test out there due to Nvidia’s optimizations for this test, it may prove to be valuable since we are not running a head-to-head competition. My main goals in running these tests were to get a general feel for a baseline set of scores, and then use them to validate and push the memory and speed settings to increase performance.

First, I started off with everything at stock settings (unless otherwise noted, all tests run at resolution settings of 1024 x 768). For all tests run, I kept the antialiasing set for “by application” and varied the anisotropic filtering through the available variables. Here are the results of the benchmark runs, grouped by setting parameters versus the three benchmark types.

Palit Graphs 1

As you can see in the chart, the primary variable in the baseline benchmarks were the anisotropic filtering. As this was enabled, the performance took a bit of a hit with anisotropic filtering enabled – 2.8% at the 2x setting, 4.9% at 4x, and 6.8% at the 8x settings for the 3dMark2001SE score. The drop in results were similar for the 3Dmark2003 scores. However, the UT2003Demo flyby scores took a major hit with anisotropic filtering enables – 11.2%, 14.0%, and 15.6% respectively. The UT2003Demo botmatch scores differences were statistically unchanged.

Next, I used the Coolbits hack to modify the settings. I turned off the anisotropic filtering and kept if off for the remainder of the testing. For the memory and core speed overclocking, I started out bumping up the core speed and memory on successive groups of video tests, starting out with core speed jumps of 7-10 Mhz and even greater memory speed jumps.

Palit Graphs 2

As you can see, some interesting things were revealed. As I began to bump up the core and memory speed, the scores quickly made up ground and showed a stable and steady progression up to a point. One point ot note - the UT2003Demo “flyby” and “botmatch” tests are both average framerate results; the botmatch is more dependent on CPU power and thus appears to have little change throughout the tests run for this card. The “flyby” is more reliant on raw video card performance and is thus more relevant for testing purposes.

At one point though, the scores began to drop and fall at about 350/717 Mhz. Oddly enough,the 3dMark2001SE score took a huge dip at the 361/732 Mhz settings but rebounded to increase again to elevated levels. I thought that I had perhaps written a number down wrong, but I ran the test again and got a similar result. I’m not exactly sure why this dip in results occur at this particular memory setting. After seeing the results fall, I tinked a bit and came up with a “best fit” setting of 340/698 Mhz which seemed to be the most stable setting with strong results, and this is the level that I’ll leave it at. Keep in mind that maybe this was just a quirk of my system…I stopped at 340/698 but the card actually ran well at a far greater overclock (see the second chart for levels achieved). No matter how high I pushed it, it still worked fine to the naked eye.

It may be strange – but in running nearly 50 benchmark tests on this video card, I never saw the first artifact. There were times, particularly in the 3DMark2003 test, where the frame rates fell painfully low – at times (especially with the anistropic filtering set to 8x), it averaged 1.5 to 2 frames per second. But once again, I never did see any ghosting or artifacts, only a slowing of framerate. Here are some screenshots of the UT2003Demo, all of which were taken at 1024 x 768.

Screenshot 1     Screenshot 2

Next, let’s talk about some subjective impressions. I downloaded the “Dawn” demo directly from Nvidia. Yea verily was I impressed..the detail was extraordinary. Despite the beauty of the fairy, I was most impressed by the wings and the gossamer effects involved. I wish that I could have easily obtained screenshots form the Dawn demo.

I might mention here that some of the Nvidia cards have a reputation for having rather loud fans, particularly the 5800 FX series. This was NOT a problem with the Palit Daytona GeForce FX 5200 Ultra card. I have a CPU fan (of course), the power supply fan, an internal fan, a northbridge fan, and an exhaust fan – and the video card fan could not be heard over any of them. I don’t know what the decibel level is because my meter doesn’t go that low - but in any case, it is very quiet.

Conclusions:

I think that Palit Daytona has come up with a card in the GeForce FX 5200 Ultra that would work for most games and applications. The results I obtained were about 10-15% below the scores that I had hoped to reach in my mind before I started the test, but I probably had unrealistic expectations. My computer is admittedly hampered and takes some performance hits from a substandard speed hard disk system and “lower-than-most-serious-gamers” CPU performance. I was impressed by the fact that the results I did get were very linear in nature, and seemed to be predictable based on the settings. This leads me to believe that an improved base system would highly correspond to an improvement in video performance for this card.

Finally, I think there needs to be some consideration of intangibles. This card will be one of the most documented and researched video cards on the market, so support and updated drivers are easily available for any system for which it is appropriate. Pricing on the internet typically runs in the range of $125-130, so it’s affordable for the level of performance. Casual gamers will most likely be satisfied with the level of performance of this card. And, with DirectX 9.0 support, the feature set is there to be useful for a significant time in the future while the games and applications catch up. I wish the included software package would have been a little more generous, but hey, anything they would have included would have been available on the internet anyway, so it would only be a time saver. With the S-video port and included DVD software, this card could easily be utilized and ported to a big screen TV. In short – a solid but not stellar performer.

(Editor's note; I was very impressed with the Palit Geforce FX 5200 Ultra. With a 3DMark2001SE score of 10,700 on my system (EPoX 8RDA+, AMD XP@2340, 512 MB PC 3200) the performance exceeded my expectations for a budget card. If you are building a budget box this is a good choice, unfortunately if you are a Geforce 4 Ti, or a Radeon 9XXX owner looking to upgrade your video this is probably not the card you are looking for.)

Pros:

  • Easy to install
  • Info/drivers widely available
  • Stable, predictable performance
  • The sweet purple color matches my black light setup in my case!
  • Able to reach a SIGNIFICANT overclock
  • DirectX 9.0 ability

Cons:

  • Somewhat spartan package bundle
  • Results lower than hoped for but better than expected
  • Experienced some very slow frame rates on some demos (but no artifacting)
  • Maybe a bit pricey in the $125 range

I would like to thank Shunny Shen from Palit for providing us with this card.

Added: August 4th 2003
Reviewer: Insulglass
Score: 8  
Related Link: Palit
Hits: 57416
Language: english

  

[ Back to Reviews Index | Post Comment ]

Reviews ©
 

Copyright © 2001 - 2014 by Jim Adkins


Distributed by Raven PHP Scripts
New code written and maintained by the RavenNuke™ TEAM


(Original PHP-Nuke Code Copyright © 2004 by Francisco Burzi)
Page Generation: 0.16 Seconds

:: fisubice phpbb2 style by Daz :: PHP-Nuke theme by www.nukemods.com ::
:: fisubice Theme Recoded To 100% W3C CSS & HTML 4.01 Transitional & XHTML 1.0 Transitional Compliance by RavenNuke™ TEAM ::

:: W3C CSS Compliance Validation :: W3C HTML 4.01 Transitional Compliance Validation :: W3C XHTML 1.0 Transitional Compliance Validation ::