GeForce 3 Ti 200: NOT a bad performer!

[Prof.Wizard]

I was expecting a larger gap... Instead this card (well, from ASUS which is well-known for its quality materials) performs less than a third worse than the next-generation flagship GeForce4 Ti 4600...

Paid it €280. Haven't even overclocked it yet! Neither the AGP.

 

[NRG = mc²]

haha I tested my Viper 550 on the machine over here... truly sad...

www.rotting-energy.net/Forums/tntpower.png

 

[Handruin]

lol, wow that's a big difference...according to their marketing department.

 

[Prof.Wizard]

lol, wow that's a big difference...according to their marketing department.

hehe... the nVIDIOTS were preaching their G4 Ti was twice as fast as the previous generation. Of course, this benchmark tests the raw computation power of the graphics chip and not the advanced features (such as the AccuView Anti-Aliasing) found only in the G4 Ti...

However I don't care... at my local reseller, the ASUS V8200 Ti 200 "Deluxe" (I've got the "Pure" model) costs €330. The V8460 costs a whopping ( ) €660...

Twice the price, but not twice the performance. Screw them... :evil:

 

[Mercutio]

Can someone with a decent internet connection and no moral shame at visiting the nvidia pages try a couple non-nvidia cards on that? I'm interested in seeing what their tools says about Matrox, 3dfx and ATI cards in particular.

 

[Pradeep]

http://vdpc.org/nvidia.gif

Using an ATI AIW Radeon 7500. The GF3 was 137%. I'm not too worried I bought the AIW more for it's TV/recording capabilities then to play Doom3. Oh and the radio frequency remote too

 

[Pradeep]

http://www.nvidia.com/view.asp?IO=analyzer

Oh that's the link to the test. Do you think if we butter up Tony he will give us all edit rights? At least say for 5 mins after we post?

 

[Handruin]

Looks like I need an upgrade. :cooler:

 

[James]

Yeah - I tried running it on my laptop, and it just gave up and told me to buy a laptop with a GeForce2Go on it. :roll:

 

[Handruin]

Yeah - I tried running it on my laptop, and it just gave up and told me to buy a laptop with a GeForce2Go on it. :roll:

The interesting part is that the GForce 4 is more then 2x what I paid for my radeon 64DDR a year or more ago and it's only 2x as good according to their charts.

That's not very impressive considering my 64DDR still plays all the games I want. (although Warcraft III is showing some stress)

 

[Pradeep]

Yeah - I tried running it on my laptop, and it just gave up and told me to buy a laptop with a GeForce2Go on it. :roll:

LOL same here. I would have thought they would be pushing the GF4 2 Go tho.

 

[CougTek]

Their related survey about user appreciation of their puny Analyser is laughtable. Both their last questions start by "What did you like..." What about what I didn't like?

Pathetic way to show their POS Analyser receiving only good comments. I wrote them what I didn't like and if they aren't happy, they can wipe their *** with my comments for all I care.

I got similar result as P.W., quite normal considering that I too own a GF3 Ti200 from Asus.

 

[time]

Prof,

That's a steep price to pay for a generic card We can sell Gainward GeForce3 Ti 200 here for about €200 plus tax

Frankly, I think they're at least as good if not better than Asus.

I'm hoping the GF3 Ti 200 will come down in price as NVidia pushes the GF4. It already looks like a much better buy than the MX 440 et al.

Has everyone checked out the relevant Anandtech reviews on this subject?

 

[Prof.Wizard]

Prof,

That's a steep price to pay for a generic card

ASUS is not a generic brand... are you serious?

And it was almost 20% more expensive than the rest of brands (ie. Creative, etc.) present at my retailer.

 

[Handruin]

Prof,

That's a steep price to pay for a generic card

ASUS is not a generic brand... are you serious?

And it was almost 20% more expensive than the rest of brands (ie. Creative, etc.) present at my retailer.

The truth of the matter is that all Nvidia cards are generic. Think of it in terms of ram, and how we distinguish "generic" ram. Generic ram, is ram made by an independent party, but uses another manufacturer's chips...this is how all nvidia based cards are made.

 

[Santilli]

What a joke..

Kyro 2, 64 mb of ram, rates:
658% on G-4 whatever:

GEO MMX 200 rates 165%. However, having another computer with that card, in the past, the Kyro 2 kicks it's butt, but, that's just my opinion.

Kyro 2 is a VERY good business card, by the way...

Have to run this test on my 5500.
gs

 

[Groltz]

Re: What a joke..

I went over to NVidia's site to try my Visiontek GeForce3 Ti200 against the performance analyzer. For some reason it would only compare my results against a GeForce 3. It showed me with 100% and a GeForce 3 with 101%. I couldn't find any way to select a different NVidia chip to compare against.

The kicker in all of this is that GeForce3 Ti 200 cards are clocked at 175Mhz core/400Mhz memory and original GeForce3 cards are at 200Mhz core/460Mhz memory. Before going to NVidia's site I used RivaTuner to overclock my card to 221Mhz core/500Mhz memory. (Running a TT Crystal Orb with Actic Silver III for cooling). That analyzer "ain't quite right".

-Steve

 

[Pradeep]

LOL, the POS is probably looking for the chipset ID and giving out a standard set of numbers. No doubt that is why it got confused with the laptop chipsets.

 

[Prof.Wizard]

Most probably there's no actual testing. They have an enormous database with all various combinations and give the results.

However, is this changing the objectivity of the benchmark? They must have conducted the tests another time... why shouldn't they be right? (provided everyone runs normally clocked, normally set video cards)

 

[Handruin]

I don't think they benchmark either. i ran it again and achieved the same results. I also watched how much CPU was used during the test and it was only 6% being utilized.

I realize the video card has a GPU of sorts, but I would think the CPU should be used a little more then 6%.

 

[Prof.Wizard]

I didn't mean they do now the benchmarks on every PC accessing the service. They have already done them in their labs and placed the results in an accessible (by the web, via the nVIDIA Performance Analyzer) link to anyone who wants to have an objective (see: nVIDIA-biased) comparison of his/her graphics card...

 

[Pradeep]

Instead of building such an elaborate cherade, why not just put up a table of all the chipsets and how they compare? It the prog is not checking the true memory performance (like the diffs between the diff Geforce4's) then it's a worthless piece of poo poo.

 

[Handruin]

So in light of this benchmark utility, what GForce brand is good to buy, and in that brand, which model?

I've had enough frustration these past few days with my radeon that I'm looking at a gforce card.

I may be a bit drastic, but I'm a little annoyed with ATI drivers right now in XP. I can't get good frame rates and I've tried tons of different drivers and settings.

 

[Prof.Wizard]

ASUS seem to give nice results. Anyone else having a G3 Ti 200 post your results for comparison!

I guess a GeForce 4 MX is right for any job you might need, Handruin, including games. And it doesn't cost much...

 

[Mercutio]

So in light of this benchmark utility, what GForce brand is good to buy, and in that brand, which model?

I've had enough frustration these past few days with my radeon that I'm looking at a gforce card.

I may be a bit drastic, but I'm a little annoyed with ATI drivers right now in XP. I can't get good frame rates and I've tried tons of different drivers and settings.

Come toward the light!

What is it that's giving you lousy frame rates? What's a "lousy framerate"?
My experience with the 7000-series cards has been excellent in 98 and 2000.

 

[CougTek]

And with the new driver (v7.68 ) I linked on the front page, they are now even better :mrgrn:

 

[Pradeep]

How did the upgrade go for you? Did you un-install the previous drivers first? Or just put them over the top?

 

[CougTek]

How did the upgrade go for you?

I haven't installed them yet because my sole computer with a Radeon card is used by someone else (I passed it to a relative who needed it). But if the 7.68 is like the 7.60 (last version I installed on my foreign system), everything should be seemless.

Did you un-install the previous drivers first?

That's how I usually upgrade drivers (in safe mode when possible too). It always works fine like that.

Or just put them over the top?

But everytime I ask my relative to upgrade the driver, that's what he does (and always with at least 12 programs running in the sys tray too). However, he didn't have any problem with that method when he installed the previous version of the ATI Radeon Win2K driver, so I guess ATI thought about this type of user when they design their drivers.

I am awared of the performance boost at higher definitions because I read a comparative review about the v7.60, v7.67 and v7.68. Nothing trully scientific, but still revelant I think. It's somewhere in Anadtech's web news section.

 

[Handruin]

In my previous upgrade to the driver set right before the one you posted today, I had major issues...

I uninstalled the old driver, then I installed ATI's latest release drivers and when i rebooted my machine went to hell. It froze every time it made it to the user selection menu and I had to press the reset button. Safe mode worked ok, but every time I uninstalled the latest drivers and reinstalled, the same BS happened. My machine continued to lock up at the user selection menu even when I reverted back to the older set of drivers.

I forget which pattern finally fixed my problem, but it took almost 4 hours of constantly hitting the reset button and going into safe mode. I'm a little hesitant to upgrade again after this, but I may try tonight.

My next issue is with refresh rates in games. The picture is never centered on my monitor because every game used to run at 60 or 75 Hz refresh. This should be fine, but for some reason the picture is 1/3 off the screen to the right. Adjusting it by using the monitor menu doesn't even fix it all the way.

So I downloaded a utility called Radeonator. This has fixed my refresh issue because I set every resolution to use 85 Hz as the minimum. For some reason the ATI Radeon 64 DDR and my IIyama VM Pro 450 do not like any other combination. I didn't have this issue with my Asus 3400 TNT card.

Another weird issue is that Counter Strike 1.3 runs much better in D3D then in OpenGL. This is quit the opposite compared to the past. Counter-Strike and half life ALWAYS ran better in OpenGL for me. I can't remember the last time I played counter-strike, so perhaps it is windows XP causing the problem.

Lousy frame rate is 15-20 FPS at 800x600 @ 16 colors in OpenGL with 1-3 people in a CS game. Warcraft III chugs like you wouldn't even imagine during large fights. I run WCIII at 800x600 16 bit color and the lowest details in all categories. It's pathetic for a 1.2 GHz machine to chug in counter-strike, I never had this problem before, and it even played fine on my PII 350 MHz.

I recently formatted my drive and reinstalled XP Pro clean, no upgrade. Is it my drivers somewhere else? I looked up information regarding BIOS setting and adjust them accordingly.

I feel as though gaming is slow on my machine right now and I'm not sure where the answer is hidden. Is it poor drivers by ATI, is it something I configured incorrectly, was it due to me install VIA 4 in 1 drivers? I have very little running in the background when I play games, and yes, I do stop Genome.

 

[adriel]

I was expecting a larger gap... Instead this card...performs less than a third worse than the next-generation flagship GeForce4 Ti 4600...

On a 1.3 cpu. But after a certain point when using a GF3 the frame rate will not really increase with a faster cpu, while with a GF4 there would still be some gains to be had beyond this point with a faster cpu.

well, from ASUS which is well-known for its quality materials

Really? I thought they used normal consumer grade quality materials.

 

[Sol]

On the debate about generic versus name brand cards. As far as I've seen there are two main differences between brand name cards from amnufacturers such as Asus and MSI and generic cards from manufacturers like sparkle.

First is that brand name cards generally seem to overclock better. This isn't always true but the odds of getting more overclockable RAM especially generally improve.

The second thing is the waranty. This varies alot between brands. Most of the generic brands have short waranties(although the visionTek G4TI cards seem to break this trend a 6 year waranty) which are voided with the drop of a hat. Of course this is true of some brand names as well. My own experience with Creative Labs being a case in point where my Geforce2 Ultra card started falling apart when I took it out of the box, a closer look revealed that the waranty on Creative Labs products pretty much excludes any circumstance thier lawers could think of (a notable clause also points out that they exclude any circumstance they havn't thought of which they are legally entitled to exclude). Fortunatly I had purchased the card from Tony and his gentle persuation, read threats of legal action, convinced them to make an exception.

I suppose the point is that a brand name product is not always better, but I think it usually is. Especially if you find a good brand which other people have been happy with.
One such brand for me has been Hercules. Having just been through my Creative experience when purchasing this card I imediatly took note of the warranty information. It contains no clauses which even mention vioding the waranty (An overclocking tool is included whch does not void the waranty) and the last clause basically says that if anything happens to the card which we havn't already mentioned we'll deal with it. I've never actually had to put this document to the test, but at the very least its nice to know that they're pretending to be on my side.


One final point I would like to make is that if any GForce card could be said to not be generic, it would have to be the Asus cards as they are the only company I know of at least who have not used the straight Nvidia reference design. I don't know if that is true for all or only some of thier cards however.

 

[Prof.Wizard]

I will never buy anything if it's not branded. I have a bad experience from old, stupid no-name products with void warranties and non-updateable drivers...

In today's competitive and complex market, support is what counts.

And to return to our main topic: Please post your benchmark results if you have a GeForce 3 Ti200 to compare the performance across different brands...

 

[time]

Please post your benchmark results if you have a GeForce 3 Ti200 to compare the performance across different brands...

Why? Numerous websites have already done this. Gainward Golden Samples are fastest because they are overclocked out of the box. Gainwards in general overclock very well so they win a lot of comparisons.

If you don't overclock and your memory clock is up to speed, you can expect identical performance out of any GF3 Ti200, whether it be Sparkle or Pine or whatever. Buy the Asus if you want 3D goggles or something similar, otherwise don't waste your money.

Although as I've said, there are certain brands I'd avoid.

 

[Prof.Wizard]

I think there's nothing better that nVIDIA's own database and benchmarking system... if something is NOT biased regarding the results, then that's nVIDIA herself!

You don't happen to have a Gainward adapter, do you? Are you afraid to post your results, time...? :wink: