3dfx Voodoo5 5500 - Conclusion
(Page 2 of 2 )
Ah, the true challenge to separate the men from the boyz, it's put up or shut up time for the V5500, benchmarks.
We're gonna do the standard benches here, and it ain't cause we're too lazy to be creative! It's because we want to use normal benchmarks that you can replicate at home to compare your own system to.
TEST SYSTEM: PIII 600e @810mhz 128mb PC 100 Corsair Ram Diamond MX300 Sound Card Generic 48x CD Rom HP Burner USR 100mb NIC Card | VIDEO CARDS: Voodoo 5 5500 Guillimont GeForce256 SDR Herc GeForce 2 32mb Vision Tek 64mb Geforce 2 |
I see no reason why we shouldn't start of with old faithful. Quake 3, we'll start low and work our way higher just for fun. Oh yeah, if ya gots beer, now would be a kickass time to break one out , it makes looking at bar graphs A LOT more entertaining.
Just to let cha know so you can replicate these exact benches, here are the game settings I used for the following benches:

FASTEST:

As you can see, the Voodoo has a hard time competing at the fastest setting, so if you are into ultra low resolution game play, this may not be the card for you if speed is of the essence.
NORMAL:

While the Voodoo 5500 starts to stack up a bit better, it is still lagging a bit behind in the normal setting. Sure, someone would be hard pressed to call 100FPS+ unplayable, the bottom line is that it is still showing to be slower than the GeForce 256 32mb SDR card which can be bought for less than half the price of the 5500 right now.
HIGH QUALITY:

Now here's where we start to see the 64mb of the V5500 catch up to the pack. Whether it's because the CPU now becomes our bottleneck or because this card shines at resolutions of 800x600 and more we don't know.
Go ahead, hit that next button and see what happens when we crank up the resolution and throw some 3dMark2000 at this thing.
Q3 Benches Cont.....
Alright, now its time to see if we can make this thing come to its knee's, while this may be a great trait in a date, it's not what we look for in a graphics board. On with the benches..
HQ 1024x768

Here we see that the Voodoo is indeed making its move. It has passed up the GeForce 1 AND the 64mb flavor of the GeForce 2. At over 80FPS this is still VERY playable.
HQ 1280x1024

At this high of a resolution, the GeForce256 drops completely out of the picture, yet the 32mb Herc still reigns king. Still, we're over 50FPS for out Voodoo5500 and THAT ain't to shabby for. Maybe not screaming fast, but playable non the less.
HQ 1600x1200

Here again we see that difference between our Voodoo and our GeForce 2's is negligible. All three are still respectably above 30FPS at a resolution that has long been considered impossible for any smooth play at all. While I personally would never play a game with just 30FPS, it wasn't long ago that we considered over 30FPS at 640x 480 smokin!!
Throughout these benches you will notice that the 32mb GeForce GTS beat the 64mb version, just to let you know, when overclocking both of these cards, the 64mb version leaves the 32 in the dust. Check out our reviews of those respective cards for the scoop on that, but this is the V5500 review and I don't want to pollute the water.
But speaking of overclocking, lets look at what this card is capable of with a little extra juice. The below scores were achieved with the Voodoo 5500 at 190mhz, which is the highest I could get this card stable. This card runs stock at 166mhz, so we were only able to squeeze an extra 24mhz from this bad boy, but hey, we take what we can get around here!
OVERCLOCKED RESULTS:
SETTING | NORMAL | OVERCLOCKED |
Fastest | 145.2 | 148.4 |
Normal | 135.1 | 136.8 |
HQ | 122.9 | 129.5 |
HQ 1024x768 | 81.6 | 94.4 |
HQ 1280x1024 | 51.2 | 57.9 |
HQ 1600x1200 | 32.1 | 37.7 |
Quake 3 Scores in Frames Per Second
As you can see, we were able to get a little better performance out of this with overclocking, but not so much that I would consider running this OC'd all the time. This card gets HOT at default speed, so it's almost crazy to push it this hard all the time for such a minimal increase. One thing that was a BIG drawback for me with this card, no matter what I tried I could not get this card stable over 90mhz AGP bus speed. So if you are into heavy FSB clocking, especially if you are still on a BX board, this may NOT be the video card for you. That said, moving on.....
Let's take a look at 3dMark2000, FSAA performance, and wrap this baby up...
We've seen what the card can do with Quake 3, let's gander at what this card is capable of achieving when up against MadOnion.com's 3dMark 2000.
3DMARK 2000 SCORES:

Ouch, I think we've found the 5500's biggest weakness. 3dMark2000 is highly feature dependant, and our V5500 is not exactly a champ in the T & L department, as is glaringly obvious here. Whether these benchmarks mean anything is subject to controversy, but one thing can't be argued, and that's that the Voodoo 5500 does NOT compete well with the GeForce 2 GTS's in this category.
Ok, the last thing we want to look at the FSAA performance. We showed you earlier how unbelievably good FSAA looks, but everything comes at a price. The price you pay for the superior graphics offered by FSAA, is performance. What we are going to show you are two Quake 3 benchmarks, each with FSAA off, 2x, and 4x on
Quake 3 FSAA SCORES:
RESOLUTION | FSAA OFF | 2X FSAA | 4X FSAA |
800x600 | 122.9 | 69.9 | 34.8 |
1024x768 | 81.6 | 44.1 | 19.4 |
If you had any questions of the cost of FSAA, there's your answer in plain language, FRAMES PER SECOND! Now one thing I wanted to point out here, these were taken running Quake 3, which is a VERY demanding game. I have played Nascar Heat at 800x600 with FSAA at 4x and it ran beautifully. So it really comes down to what you prefer, if it's blinding speed, then FSAA is not for you, but if graphics quality is high on your list, you'll never play another game without it.
CONCLUSION:
This is actually a tough one, on the level I would have to say that this is a very strong card and certainly would suite any gamers needs. But for the $249 price that this card is going for on Pricewatch right now, it's hard to recommend this card over a GeForce2 32mb that is running a bit less. So I am leaving the conclusion up to you on this one. You've seen the scores, you've seen the FSAA, you've seen the performance. So really, all that's left is for you to get intimate with a six pack of your favorite beer and see which card you like best. Maybe take a close look at your wallet and see if wants to shed some weight, if so then maybe this card is for you.
One thing that I learned from all of this is not to go into any review thinking you already know the outcome. When I began writing this I was positive that the GeForce2 was going to treat the Voodoo5500 like an ex-girlfriends step child, I was mistaken. The Voodoo 5 5500 held it's own at the resolutions that we gamers care about, 800x600 and above. It even took a bit out of the GF2 in a few of the benches, which more than surprised me. I leave this review with a "what if" floating around in my mind. What if 3dfx had chosen to go with DDR instead of the less expensive (at the time) SDR? If the 5500 hangs this close with its 64mb of SDR, how different would these benchmarks have been if it were DDR? I guess that's a question we'll have to wait for, at least until the new card 3dfx is working on comes out.
PRO'S:
Second to none in graphics quality, FSAA pays off.
Molex connector on board to provide its own power.
Meanest looking card on the market (IMHO)
CONS:
Can only handle 90mhz AGP Bus speed. (at least this one)
Price is a bit high for the performance ($249 OEM) ($300 Retail)
Gets tore up at low resolutions by the competitors
With no hardware T&L, not too impressive in 3dMark2000
| DISCLAIMER: The content provided in this article is not warranted or guaranteed by Developer Shed, Inc. The content provided is intended for entertainment and/or educational purposes in order to introduce to the reader key ideas, concepts, and/or product reviews. As such it is incumbent upon the reader to employ real-world tactics for security and implementation of best practices. We are not liable for any negative consequences that may result from implementing any information covered in our articles or tutorials. If this is a hardware review, it is not recommended to open and/or modify your hardware. |