View Full Version : NV30 benchmarks?
PreservedSwine
08-21-02, 03:46 PM
Originally posted by robert60446
I lost faith in Ati, after “lazy drivers” case. Hardware can’t work without good software…Ati proved already, that they can’t even get close to nvidia drivers. Customer service and support, that’s the way to go my friend…
I agree drivers are important, but also remember nothing lasts forever.......
Ask anyone w/ an ATI R8500, they're some of the happiest gamers on the planet:)
And no need to tell me about lousy drivers, I owned an NVIDIA TNT1- talk about CRAP!
My point is that although Nvidia seems to haver a good reputation for drivers, it hasn't always been that way. In reality, they are currently no better than ATI's. In the last 12 months, ATI has made considerably more driver releases than Nvidia. The stabiltiy has been excellent, and now the R8500 is darn near the Ti4200, and surpassing it once AF is enabled. Althogh, FSAA turns things in favor af Nvida as well.
just my .00000002
:)
mcortz_2000
08-21-02, 04:25 PM
How can u talk about drivers that were out way back when, that was a long time ago. The driver quality of recent cards is what has determined peoples opinions, and Nvidia rules compared to Ati. Ati's drivers do not compare to Nvidia's driver. They have had far more problems and i am talking from experiences and many other people have had the same issues............
PreservedSwine
08-21-02, 04:33 PM
Originally posted by mcortz_2000
How can u talk about drivers that were out way back when, that was a long time ago. The driver quality of recent cards is what has determined peoples opinions, and Nvidia rules compared to Ati. Ati's drivers do not compare to Nvidia's driver. They have had far more problems and i am talking from experiences and many other people have had the same issues............
Because someday, "way back when" will be desribing today, and you need to know the differnece when things are changing before your very eyes.
My point was that just becuase ATI had lousy drivers once (RAGE), it doesn't mean they always will. Just as Nvidia changed their lousy reputuaion....
The Original Radeon, although upon release it needed a few tweaks had excellent drivers.
The R8500 original drivers needed improvement as well- AND LOOK AT WHAT HAPPENED. ATI has really cleaned up it's driver image. The R9700 seem to continue that trend. It's one of the best new Video card reviews (in relation to driver stability and options) I've ever seen.
Nvidia drivers have been good as well, but perhaps you have forgaotten about the Geforce3 release? Utter crap. As bad or worse than the R8500 initial drivers.
Both companies, IMO, put out very competive, good drivers today.
mcortz_2000
08-21-02, 04:40 PM
Dude saying that the 8500 had even decent drivers is really far fetched, i had a 64MB retail 8500 and they were not nearly as good as u make them out to be(or were). The current Nvidia drivers are far from perfect but for the most part have had far fewer incompatibility problems then Ati cards....
Bigus Dickus
08-21-02, 04:56 PM
Originally posted by mcortz_2000
Dude saying that the 8500 had even decent drivers is really far fetched, i had a 64MB retail 8500 and they were not nearly as good as u make them out to be(or were). The current Nvidia drivers are far from perfect but for the most part have had far fewer incompatibility problems then Ati cards.... So you're comparing the 8500's release drivers with current NVIDIA drivers? Fine, I'll compare the GF3 release drivers with the 8500's current drivers and conclude that NVIDIA has crap drivers. Fair? What's the difference?
I think PreservedSwine made a good (probably the most important) point in this whole thread... the 9700 reviews are the best I've seen of any new graphics card release that I can remember (neglecting psudo-new cards like the GF2 Ultra or the GF3 Ti). There were very few and relatively minor driver bugs found (in fact, most reviews found none, and just commented that ATi had provided a list of known bugs). Talk about changing your driver reputation... this is one sure way to do it.
mcortz_2000
08-21-02, 05:18 PM
No, maybe u misunderstood me. Let me clear this up.
The initial GF3 drivers were better than the initial Ati8500's drivers.
The current GF3 and GF4 drivers are better than the 8500's current drivers.
I have not worked with the 9700 so i can make no comment on those.
Bigus Dickus
08-21-02, 06:28 PM
Originally posted by mcortz_2000
No, maybe u misunderstood me. Let me clear this up.
The initial GF3 drivers were better than the initial Ati8500's drivers.
The current GF3 and GF4 drivers are better than the 8500's current drivers.
I have not worked with the 9700 so i can make no comment on those.
Really? I got a chance to help my friend do a fresh install of everything when his new Ti4200 showed up last weekend. What did I take away from that experience?... I thank God that I have an ATI card and don't have to go through that driver shuffle game. His system is still unstable.
:)
Isolated case? Perhaps. IMO current ATI drivers are every bit as good as current NVIDIA drivers, and as far as user features, functionality, and configurability, they are far better. Initial GF3 drivers better than initial 8500 drivers? A matter of opinion I guess... I never had a problem with my 8500 (got the Dell deal immediately after they were released), and I guess the only real complaint would be the lack of Smoothvision (which, IMO, isn't any different from what the initial drivers had in a less "optimized" form and without the gimmiky name).
mcortz_2000
08-21-02, 06:38 PM
Ok firstly the stability of his system has nothing to do with quality of drivers unless something was wrong witht he card.
The fact that he an unstable system points to something else being wrong. What are his specs by the way?
I have installed countless nvidia and ati cards and the problems have been far more pronounced on the ati boxs.
Also your comment about still unstable, what was wrong before.
Anyway lets not get caught up in opinion bashing. You have yours and i have mine.
The rule of thumb is you look at everyones experiences with ati and nvidia cards and work out the percentages and so far the nvidia setups have had fewer problems, not no problems just fewer problems...
I have read somewhere ,maybe HARDOCP
that RADeon9700 is very unstable in AGP8x boards ,
anyone can confirm this ?
to soon to get conclusions but it will be better for ATI to fix that,it would be a shame that this problem cannot be fixed at all with drivers ,but with hardware revisions..
i cannot comment on ATI cards ,never needed to switch to ATi
side ,because since Tnt2 i have never had any problems with
drivers in Nvidia card.. in the other hand i remember my friend
changed his radeon2 for a geforce2 ti ,because he was tired of
hunting for leaked drivers to fix ,its drivers issues with
in windows 2000..
Maybe Ati has now more mature drivers ,but Nvidia has proven
that can deliver more performance in lower harware..
remember Geforce3 (detonators driver) vs Radeon2 ..;)
PreservedSwine
08-21-02, 08:52 PM
Originally posted by Nv40
I have read somewhere ,maybe HARDOCP
that RADeon9700 is very unstable in AGP8x boards ,
anyone can confirm this ?
to soon to get conclusions but it will be better for ATI to fix that,it would be a shame that this problem cannot be fixed at all with drivers ,but with hardware revisions..
i cannot comment on ATI cards ,never needed to switch to ATi
side ,because since Tnt2 i have never had any problems with
drivers in Nvidia card.. in the other hand i remember my friend
changed his radeon2 for a geforce2 ti ,because he was tired of
hunting for leaked drivers to fix ,its drivers issues with
in windows 2000..
Maybe Ati has now more mature drivers ,but Nvidia has proven
that can deliver more performance in lower harware..
remember Geforce3 (detonators driver) vs Radeon2 ..;)
Yes, I've seen report of the AGP8x slots being unstable, but it's a bit too earlt to tell why..
Funny you mention Geforce 3 det. to R8500--look at them now- the R8500 is signigigantly faster than ALL Geforc3's, with the excpetion of the Ti500. The Ti500 can hang pretty close, but the R8500 is still the better performer.
So what was your point about drivers again?
Comparing a mature driver set to a brand new card and drivers? Funny, cause that's exactly whats going to happen on the NV30 launch- ironic, huh?:)
jbirney
08-21-02, 09:41 PM
I have owned a ATI 8500 since last Nov. No real big issues no stability issues. I also develop Mods for UT/Q3 based games and had no issues in developing those with the 8500. Guess I got lucky?
StealthHawk
08-21-02, 10:12 PM
Originally posted by PreservedSwine
Yes, I've seen report of the AGP8x slots being unstable, but it's a bit too earlt to tell why..
Funny you mention Geforce 3 det. to R8500--look at them now- the R8500 is signigigantly faster than ALL Geforc3's, with the excpetion of the Ti500. The Ti500 can hang pretty close, but the R8500 is still the better performer.
So what was your point about drivers again?
Comparing a mature driver set to a brand new card and drivers? Funny, cause that's exactly whats going to happen on the NV30 launch- ironic, huh?:)
yes, the R8500 is a lot faster in selected games and benchmarks. but for "some reason" it falls behind GF3 in many DX games :rolleyes:
conveniently it performs better in games that are benchmarked heavily for reviews, as opposed to say Max Payne, Aquanox, UT2003, Commanche4, etc, etc
enough with the "card X is ALWAYS faster than card y" bull****. we know it isn't true. both ATI and nvidia cards have their strengths and weaknesses in certain games, PERIOD.
PreservedSwine
08-21-02, 11:37 PM
Originally posted by StealthHawk
yes, the R8500 is a lot faster in selected games and benchmarks. but for "some reason" it falls behind GF3 in many DX games :rolleyes:
conveniently it performs better in games that are benchmarked heavily for reviews, as opposed to say Max Payne, Aquanox, UT2003, Commanche4, etc, etc
enough with the "card X is ALWAYS faster than card y" bull****. we know it isn't true. both ATI and nvidia cards have their strengths and weaknesses in certain games, PERIOD.
You are kidding yourself- although there is some truth to what you say about cards relatively equal in some areas, some have strengths and weaknesses- but pls- R8500 is superior in nearly ALL application over ALL Geforce3 cards.
Sharky latest Video card round-up will show what a beast the R8500 has become-It is serious competition for the Geforce 4 Ti4200
http://www.sharkyextreme.com/hardware/videocards/article.php/1449721
Hope that helps:)
StealthHawk
08-22-02, 12:24 AM
yes, i already thought i acknowledged that the R8500 was definitely faster in many cases(read: quake3 based games mostly), but not all cases. building such a generalization from scores given from 4-5 benchmarks are just that, baseless generalizations. i will look for benchmarks - later. maybe things have changed even more in the past few months though ;)
edit: benchmarks inlcuded
http://www.anandtech.com/showdoc.html?i=1647&p=6 take a look at benchmarks from UT2003. GF3ti500 is right on par with R8500, despite the R8500 being the technically superior chip.
http://www6.tomshardware.com/graphic/02q2/020418/vgacharts-01.html#aquanox the drivers being used might be somewhat outdated, as this was from April. again the ti500 is right on par in Commanche, and leading strongly in Max Payne. it isn't too far behind in jk2(-5fps), although i'm not sure that jk2 is using highest quality settings, and the R8500 would certainly increase its lead as resolution increases.
there used to be a link to a R9000Pro review that had gf3 scores in it, but it looks like it went down the drain when the forum data was wiped...that had some more games in it, if i remember correctly.
robert60446
08-22-02, 10:49 AM
Originally posted by PreservedSwine
Yes, I've seen report of the AGP8x slots being unstable, but it's a bit too earlt to tell why..
Funny you mention Geforce 3 det. to R8500--look at them now- the R8500 is signigigantly faster than ALL Geforc3's, with the excpetion of the Ti500. The Ti500 can hang pretty close, but the R8500 is still the better performer.
So what was your point about drivers again?
Comparing a mature driver set to a brand new card and drivers? Funny, cause that's exactly whats going to happen on the NV30 launch- ironic, huh?:)
Preservedswinie, wakeup man! Welcome to the real world…you should take a correct pill. Look: “…In 3DMark 2001 SE, the Radeon 9000 Pro finished in third place, behind the GeForce 4 Ti4200 and the Radeon 8500.”, “…Moving on to Comanche 4, again the GeForce 4 Ti4200 takes the expected spot in first place. The 9000 Pro again sits behind the 8500, but by a very small margin. This time though, the MX440 takes the lead in 1600X1200 resolution, stealing it away from the Radeon cards most likely due to driver differences…” and etc. “Driver differences…”Ati still is behind man! Better believe it!.
SnakeEyes
08-22-02, 12:20 PM
8500 release drivers were as good as the release drivers for the first GF3? B.S.
From what I've seen posted everywhere by owners and in reviews of older cards, and more recently from the 9700 Pro reviews, ATI's finally getting a handle on driver BUGS, but still is trying to get the performance optimizations down. As far as the 9700 goes, at least for all the benchmarked games, the drivers are looking good for the initial release, feature/bug-wise, but they're still in need of optimization, as based on the fact that it's beaten when CPU limited by GF4 in many places. That shouldn't happen, but isn't a severe concern, after all it IS on par with GF4 in those situations, and still blowing it away when the GF4 becomes the bottleneck.
As for the GF3 - 8500 comments, I had an original GF3 right after they came out. The drivers released for the card were far above and beyond the drivers released for the 8500 initially. Why do I say that? There were practically no bugs or compatibility problems with ANY of the games I tried on my GF3. As far as optimizations go those drivers needed work but everything seemed to function as promised out of the box. Granted there were hidden features that weren't supported, but then again I never knew anything about those when I bought the card, so it was like getting free candy once support was added. To me, the released GF3 drivers were very much like the 9700's released drivers. That wasn't so for the 8500, which only got there after several months (and if you believe some owners, still aren't there, but that's an argument for someone else).
Where does this leave the 9700? Dunno. I'll wait and see what the guinea pigs among are community have to say once they try the card out to see if it lives up to the quality all those previews and reviews are saying it has. I'm gonna do the same thing when nV3x comes out, since I also agree that TNT1 cards had poor drivers (I remember those well..), and I consider the nV3x to be like the original TNT, since it's a brand new architecture, likely requiring groundup rewrites of the drivers, versus adding new code for new features and tweaking existing code for slightly updated hardware features (which is what I think nV has been doing since TNT2, thanks to a mostly common hardware set + additions, all the way up to the GF4 line).
Kudos to ATI for making drivers that look as good as they do so far for the 9700's release, but it still hasn't passed the final test- the one where the real owners tell us their experiences (gonna be fun reading those reports and sorting the biased from the non-biased too- ugh).
Edit: Removed the word 'severe' from my characterization of the optimizations the 9700 drivers still need, in deference to bd's ego. ;)
Bigus Dickus
08-22-02, 12:39 PM
Originally posted by SnakeEyes
As far as the 9700 goes... the drivers are... still in need of severe optimization, as based on the fact that it's beaten when CPU limited by GF4 in many places.
..it IS on par with GF4 in those situations...
So if the 9700 is trailing by 5% or less in a handful of CPU limited situations, that means the drivers need severe optimization? Perhaps there's plenty of room for improvement left in the drivers, or perhaps there's only room for a 5% improvement.
We don't know either way... but what we do know is that your conclusion isn't based on logic or evidence, but simply your assumption that the initial drivers must be bad.
:rolleyes:
SnakeEyes
08-22-02, 12:54 PM
Originally posted by Bigus Dickus
So if the 9700 is trailing by 5% or less in a handful of CPU limited situations, that means the drivers need severe optimization? Perhaps there's plenty of room for improvement left in the drivers, or perhaps there's only room for a 5% improvement.
We don't know either way... but what we do know is that your conclusion isn't based on logic or evidence, but simply your assumption that the initial drivers must be bad.
:rolleyes:
Actually bd, the only thing we know is that your conclusions aren't based on anything but fanboyishness. But that's beside the point. 5% is a rather large margin for a card that is MUCH more powerful (in theory) than the generation older part. I'll give you the fact that since it's CPU limited, theoretically both should give identical results. The fact that there's up to 5% difference says that there are some driver inefficiencies that need worked out.
Oh, and please show me where I said the drivers must be bad? Reread closely. I don't consider the current performance bad at all, especially for release. Much the same as I take the arguments that the GF3's drivers were bad at release as a bunch of baloney. Poorly optimized (as far as performance went), yes. Bad, no. All functions worked as advertized, and in higher resolutions or with extra goodies on, they shined, just as the 9700 is doing now.
What I DID say was that it still hasn't passed the final test. That being the one where real owners have the card, run the games they own (and not just all those that are repeatedly benchmarked by review sites), and report their experiences. This will tell us whether ATI truly has covered the compatibility / bug issues, or whether they've just done an exceptional job of making sure that all the stuff used for benchmarking by these sites works right. There's already anecdotal evidence that at least one demo has problems (read the reviews), but I'm willing to say that one problem isn't unusual (even our GF cards had one or two blemishes here and there). It's whether there are many, and how severe, that will tell the tale.
So stop trying to put words in my mouth (or on the topic), and make logical comments, or don't comment.
Thanks for coming. :rolleyes:
Edit: BTW, I'll retract the word severe, but nothing else.
robert60446
08-22-02, 01:06 PM
Yeaaahhh Bigus Dickus, listen to SneakeEyes, he is a real Man! And stop your heresy now!
SnakeEyes
08-22-02, 01:23 PM
I dunno if I'd call it heresy. After all, the 9700 is the best card out now (or for most of us, Real Soon TM), and conceivably for the next couple of months. But twisting other posts to mean more than is said is going a bit far..
Personal stance on R9700:
1) It looks like a great card. Fixed all the issues I had with 8500 and added the features / capabilities I've been waiting for. Drivers appear satisfactory, possibly even great (jury's still out, and I personally have time to wait for it to come in, since I can't buy now anyway). A would-buy if-I-could-afford-it-now product.
2) Don't have the money right now (re: 1. above), due to the wrong time of year for my upgrade budget. Also, I have a GF4Ti4600 right now, which is 'More than good enough for anything out now', though I'd like to be able to do the couple extras 9700 adds- full aniso / 4xFSAA in all my games at high resolutions and framerates.
3) Since I've gotta wait, the nV30 WILL be out before I can buy. I can wait a little longer to see owners report on problems, likes/dislikes, performance, for both cards, then balance that against price and make my purchase. Besides, this is only smart for me, given 1 and 2.
4) Did I mention that 1, 2, and 3 have been said time and again elsewhere in these forums in past posts I've made, and most people agree, or can at least see the logical reasoning behind it. At least there haven't been any snide or negative comments about them before.
5) robert60446- you seem a bit strange. :D
Point made. Signing out. (Rain's coming down hard here too- that's also a fact, though it has no relevence to this topic whatsoever..)
robert60446
08-22-02, 01:34 PM
SneakeEyes, you are from Lake Zurich, I’m from Romeoville (IL, 22 miles from Chicago on I-55)-you are not so far from me... Some day I will buy you a beer!
SnakeEyes
08-22-02, 01:39 PM
Wow. You're starting to seem very normal all of a sudden.
So when can you get the beer? :D
robert60446
08-22-02, 01:51 PM
Don’t worry I’m normal…just only these strange voices in my head…. just joking. Sometimes I like to watch how people are ready to die in the name of some ideas…how they can manipulate and use against you, your own words…this is very interesting :D .
Matthyahuw
08-22-02, 03:07 PM
oh no, another one of you people that are too close!
I've already persoanlly met Ed (saturn), but I dunno about the rest of you ppl :eek:
Bigus Dickus
08-22-02, 04:13 PM
Originally posted by SnakeEyes
Actually bd, the only thing we know is that your conclusions aren't based on anything but fanboyishness. pfftt.
Perhaps there's plenty of room for improvement left in the drivers, or perhaps there's only room for a 5% improvement.
We don't know either way.
Conclusion... where? Based on fanboyism? Since when has declaring that we simply don't have enough evidence to do anything other than make wild speculation a "biased" or fanboy attitude? From many of your other posts, I would have thought you would agree on this point. Oh, and please show me where I said the drivers must be bad?
As far as the 9700 goes... the drivers are... still in need of severe optimizationYeah, bad is too generic of a word, I agree. Still, your wording was clear... you seem to think the initial performance of the card is much less than what it should be. My opinion is that it might be, but we simply don't know. Neither of us is informed enough about the hardware or current drivers to make such an assessment. Your "evidence" is anectdotal. Perhaps you're right, perhaps you're wrong, but I can't believe you'd refuse to see the other side of the argument because of pride, and then take a cheap shot about my "ego." Pretty sad.
So stop trying to put words in my mouth (or on the topic), and make logical comments, or don't comment.
Thanks for coming. :rolleyes:You're amusing. I guess rhetoric and cheap insults work with some people? :rolleyes: right back at you.
And for the record, I'm in the same boat as you. I spent $200 on my last graphics card, and don't intend to double that outlay anytime soon. My "planned" upgrade was for sometime early next year, at which time I'll be doing the same as you... reading reviews, reading forums, and trying to make an informed decision. I hope the NV30 is as good as it's hyped to be.
vBulletin® v3.7.1, Copyright ©2000-2012, Jelsoft Enterprises Ltd.