View Full Version : Saturday Commentary
aaahhh52
08-21-04, 12:32 PM
meow
meow
That's funny. (crazy)
Daneel Olivaw
08-21-04, 12:57 PM
I remember the good old days of tweaking my P3 650 with 192MB of sdr133 and AGP S3 Savage4 32MB.
Getting games to run well (16bit color) (ST-Armada, QuakeIIIA, UT, DeusEX) was the most fun I had. I actually had to patch together a WinXP driver for the S3 to run Return to Castle Wolfenstein (openGL). That was 4-6 years ago, nowadays I have a bit more money, and I'm not squeazing quite as much from my hardware anymore, I replaced a Ti4200 in january 2004 with a FX5900xt. I just know that when GTs will be <=300$ CAN, I'll be buying.
It's not exactly like my Ti4200 couldn't run games anymore, I just wanted to be able to stop seeing this 'moiré' in far away side-seen textures... (AF)
We're all spoiled brats :)
Given to my sister:
| P3_650@866 | 3x128MB_SDR133 | GA-6BX7 | 1x20GB_Maxtor | JATON_GF4MX440 |
(the S3 Savage4 is displayed proudly... in a small box sitting in a bigger box in a closet)
my video upgrades
PCI Ati Mach64 - Ati Xpert98 - S3 Savage4 - GeForce4 MX440 - GeForce4 Ti4200 - GeForceFX 5900XT
on the side
GeForce2 MX200 - GeForce 5700 256MB (refunded) - Some sort of Voodoo - Ati rage something
john19055
08-21-04, 01:03 PM
I was told that I would regret it if I got a 6800Ultra over a X800XT PE,since I have been with Ati for a while and the 9500pro and the 9800pro witch were great cards and I was told the I.Q could'nt be beat. So I waited and waited and waited,over three months had past and they would just keep saying you should have it in a week or less.Well I got tired of being lied to even though it was'nt there fault that ATI could'nt deliver.But after seeing more more reviews from people that have a 6800 GT or Ultra and knowing that the PS and VS 3 should help the card on some future games and a few out now.I got a Evga 6800Ultra witch will overclock past the the Extreme series,witch surprise me ,but I run it at default speeds because it is fast enough with any of the games that I play.IMO the I.Q. is just as good as the X800Pro that I had for a month and I was suprise when I ran the benchmark 2003 and got 12,000 compaired to the 5600 for my 9800pro and 8,700 for my X800pro,and can hit over 13,000 with some overclocking of the card.But benchmarks don't mean that much if they don't run a game right.But so far ever game that I have installed runs perfect with the 6800 ultra and that could'nt be said when I had my X800pro witch had problems with flashing textures in Pain Killer,but I guess all new video cards has a few problems when they first come out.
I guess I can stop rambling on but the 6800ultra has been a lot better card then I thought it was going to be,I thought that I was just settling for secound best ,but after haveing the 6800ultra and playing my games ,IMO I think I made the right decission.
SpeedDemon
08-21-04, 03:06 PM
I love how now after we all have gotten used to playing most of our games with AF and AA, when newer games come out and we have to turn these features off to get acceptable framerates on still great cards like the 9700pro and 9800pro, we just feel like something is missing. At least I can't stand jagged lines and blurred textures anymore. That just seems like a finishing touch that should be expected, and to me running a game without it just seems kind of silly unless you are playing in a tourney and need every last frame. I personally am waiting for the next series of GeForce6's in PCI-e so that I can run them in SLI and for sure get perfect performance at 10x7x32 with 4xAA/8xAF in Doom3 with never a dip below 60fps. If I can pull those kind of benchmarks then I'm sure that I'll be happy for some time to come. Until then I'm going to stay with my 9700pro which is a great card and I think that right now is one of the most exciting times in gaming with all of the great games we have and have coming (titles coming to mind; Doom3, Half-Life2, UT2004).
ppl are just more demanding now, more picky, and trying to look for more and more things to criticise. Ppl got bored for FPS, so they started IQ comparisons with 10x maginfied screenshots... Guess when the real stuff gets boring, it's time for the details. I still take the stand that if it takes a screenshot side by side to show me a difference, it's not nearly enough of a difference for me to care. I'm still all about the FPS :)
I also don't use AA but always use AF. Personal preference.
I also got out of the numbers game. If a game doesn't slow down, it's good enough for me :)
things have changed.
Also remember to check these articles:
1. http://www.driverheaven.net/showthread.php?p=423812#post423812
2. http://www.firingsquad.com/hardware/half-life2_vst/default.asp
3. http://www.vr-zone.com/?i=1193
In these articles X800XT PE is 28 to 44 % faster than 6800U in 1600x1200 4X/8X - and only Firingsquad and Driver Heaven seem to have used MAX settings in the VST test (= "reflect all"), while Tech Report only used the less intensive "reflect world" detail setting (Games Depot has unknown settings). So clearly HalcYon could seem quite biased by writing "Now I don't want to sound like an NVIDIA apologist here, but the debate raging for 70 fps vs. 60 fps at 16x12 4X/8X is hardly worth amount of time we are are wasting on it". The differences seen at the sites mentioned above far exceeds "70 fps vs. 60 fps" and I am quite surprised that NVNews has not mentioned the VST reviews on VR-Zone, FiringSquad and Driver Heaven. 6800U is doing much worse in these test compared to X800XT PE than in the tests from Gamers Depot and Tech Report and in my opinion it is only fair that the readers of NVNEWS get the whole story, unless NVNEWS wishes to conceal some potentially important aspects regarding 6800U vs. X800XT PE in Half-Life 2 VST.
ppl are just more demanding now, more picky, and trying to look for more and more things to criticise.
People should be more demanding......hell, for top-end vid they are paying $400-$500+. Reason enough me to be critical (soapbox)
kinda makes that 299CDN voodoo2 look pretty cheap doesn't it?
there was a time I wouldn't pay over 100 for a vid card.
People should be more demanding......hell, for top-end vid they are paying $400-$500+. Reason enough me to be critical (soapbox)
i agree, sort of. I demand mindblowing performance... Just that a lot of things ppl are spending a lot of effort pointing out these days I hardly care about. I can tell the diff between AA off and on, or AF off or on... but anything less obvious than that I have a hard time noticing. lol.
saturnotaku
08-21-04, 11:56 PM
it is only fair that the readers of NVNEWS get the whole story, unless NVNEWS wishes to conceal some potentially important aspects regarding 6800U vs. X800XT PE in Half-Life 2 VST.
If people browse the forums, it's quite obvious that the links you posted have been discussed ad nauseum. But I did re-read the FS comparison and there is no mention of how they benchmarked the 6800 series with regard to trilinear and anisotropic optimizations. They also did not bench with the 65.62 drivers, which has been proven to close the performance gap between the X800 and 6800.
Thank you, please drive through.
I stared at that post for about an hour, Ed, tryin to think of what to do... You summed it up nicely. :thumbsup:
I do however think it's reasonable to assume that once it's all said and done, ATI cards will have a ~10% performance advantage @ higher rez with AA and AF on...
Saturnotaku, if you took the time to actually read the postings at Driver Heaven you would know that they later did test 6800U using the 65.62 betas:
http://www.driverheaven.net/showthread.php?t=53712&page=4&pp=15
Quote: "Probably worth noting that all optimisations were on for the 65.62"
65.62 did improve 6800U quite a bit, but 6800U was getting about 52 FPS in 1600x1200 4X/8X (MAX settings) VST, while X800XT PE achieved about 72 FPS - that is a nearly 40% lead by X800XT PE. We do not know how much impact these VST scores will have on performance in the final version of HL2, but lacking better tests than the VST, I do find a 40% performance difference disturbing. So clearly 65.62 in no way closed the gap between 6800U and X800XT PE using MAX settings in the VST, as you wrongly are claiming. Furthermore this is not a discussion about getting the 6800U or X800XT PE to look best (who wins or not is not the primary issue), it is only a matter of showing all the relevant data.
So clearly HalcYon could seem quite biased by writing "Now I don't want to sound like an NVIDIA apologist here, but the debate raging for 70 fps vs. 60 fps at 16x12 4X/8X is hardly worth amount of time we are are wasting on it". The differences seen at the sites mentioned above far exceeds "70 fps vs. 60 fps" and I am quite surprised that NVNews has not mentioned the VST reviews on VR-Zone, FiringSquad and Driver Heaven. 6800U is doing much worse in these test compared to X800XT PE than in the tests from Gamers Depot and Tech Report and in my opinion it is only fair that the readers of NVNEWS get the whole story, unless NVNEWS wishes to conceal some potentially important aspects regarding 6800U vs. X800XT PE in Half-Life 2 VST.
Do you sit at your computer with a tinfoil hat on so the spy satellites can't read your mind? I didn't link Driver Heaven, it's got to be a conspiracy against ATI... :retard:
Actually, I did post a link to VR-Zone's benchmarks. I also made reference to the Driver Heaven "benchmarks" in a later post with out a link or mentioning them by name. As for the Squab, I don't like them, so I don't link them.
But you probably want more on why I did not want to link to Driver heaven's scores? First, their methodology came into question immediately when they tried to compare 6x AA performance on ATI to 8xS performance on NVIDIA. Smacks of someone stacking the deck for shock value in a graph with an asterisk. There are people replying in DH's forum with lesser systems running the exact same benchmark settings that are getting higher scores. Saturn already mentioned that DH's scores have been beaten to death in the nV Forums already. Most of all, VR-Zone and Squab's benchmarks contradict the performance gap which you so proudly tried to enlighten us poor NVIDIOTs about, unless you can make "28 to 44 %" out of 74.1 vs. 64.5.
Now to the real reason for me to reply to your post. You missed the point and made my point at the same time. The CS Source benchmark is a purely synthetic benchmark that has become the primary hardware world debate because Driver heaven has the gall to post results SO skewed one way that everyone is ignoring EVERY other site's results, you included evidently. It has little to do with how actual gameplay will be when HL2 is released because 1) no one knows if Valve is done tweaking the engine (SM3.0 patch?) and 2) there will likely be driver updates for HL2 from both ATI and NVIDIA (65.xx, 4.9 Cats).
The purpose of the stress test, IMHO, is to make sure that the engine functions properly on different flavors of hardware and all of the features on the rendering engine do as they should to produce the desired atmosphere. That's it. The usual suspects in the the usual venues have turned it into page after page of your-card-sux. Let it go. And that is directed to both the fanATItics and NVIDIOTs...
Now I'm pissed at myself, because I let you turn me into a hypocrite. Then again, the cat woke me up early and I couldn't fall back asleep, so at least I had nothing better to do...
You feel like discussing Doom3 performance? Kidding. Just a joke...
...and I am quite surprised that NVNews has not mentioned the VST reviews on VR-Zone, FiringSquad and Driver Heaven. Had they sent news of their articles to news@nvnews.net, we probably would have posted them. VR-Zone usually sends us news, but we let their second article on the subject slip through the cracks. It's been hectic around here during the past week. I've been troubleshooting performance issues with vBulletin and mySQL and Brian is filling in for our regular news poster who's taken time off to take care of personal business.
Yup...but now I am back! :D Sorted all my moving, broadband, computer problems. Things will start moing on the front page now..as far as roundups are concerened.
@rune-sr
Vr-zone's CS test was linked by BrianG...if u look back in the archives.
Then again, the cat woke me up early and I couldn't fall back asleep, so at least I had nothing better to do...
LOL. I can relate. We have two new kittens that play tag in the middle of the night. I'll post some pics in the Open Forum later on today.
saturnotaku
08-22-04, 10:25 AM
Furthermore this is not a discussion about getting the 6800U or X800XT PE to look best (who wins or not is not the primary issue), it is only a matter of showing all the relevant data.
And a lot of people who have participated in this thread (http://www.nvnews.net/vbulletin/showthread.php?t=35080) have come to the conclusion that DH's benchmarks are skewed. DH is ATI biased, they might try a thinly veiled attempt at objectivity but most of us here (and elsewhere) see right through it.
Further, it's been plainly obvious to people following this situation that ATI was going to be better in Half-Life 2. Not a single person has denied this, but we are well within our right to question one site who numbers are quite a bit different than other places who have run similar tests.
Lastly, there is more than one person who is going to wait for the final game to be released. In addition, we also have come to the conclusion that the 6800 still does very well - even 50 fps at 1600x1200, 4x AA/8x AF is nothing to sneeze.
Again, read through the thread and you'll see people have drawn their own conclusions.
Hmmm, seems I missed the link to the updated VST benchmarks at VR-Zone.
I'll link them right now. I did link the initial round of benchmarks at VR-Zone based on gameplay which Muya referred to.
OK my bad...I thought BrianG had already linked the vr-zone article
Woodelf
08-22-04, 10:33 AM
AA & AF are big issues because they are the first thing's turned off when a new (performance taxing) game comes out. Been dealing with that for a while, and will again. The next game to come out that really taxes our performance will again result in AA & AF being turned off first. Why, because that's just how important they really are. To me, a decent amount of default filtering and no AA at 1280x1024/960 are just fine.
AA & AF are big issues because they are the first thing's turned off when a new (performance taxing) game comes out. Been dealing with that for a while, and will again. The next game to come out that really taxes our performance will again result in AA & AF being turned off first. Why, because that's just how important they really are. To me, a decent amount of default filtering and no AA at 1280x1024/960 are just fine.
Thanks for getting back on topic.
I honestly believe that AA and AF are the new 32-bit color. They are taken for granted and are part of the gaming experience now. I see most game engines and game designers accounting for performance with AA and AF enabled. Hence, the games now control settings like AA and AF. To coin the ad slogan, "The way it's meant to be played."
That said, Doom3 does look great with out AA, but I think that is mainly due to the darkness of the environment.
saturnotaku
08-22-04, 11:04 AM
That said, Doom3 does look great with out AA, but I think that is mainly due to the darkness of the environment.
I mean, what good is AA if you can't see it in the first place. Hehe. (wack)
I mean, what good is AA if you can't see it in the first place. Hehe. (wack)
:retard:
vBulletin® v3.7.1, Copyright ©2000-2012, Jelsoft Enterprises Ltd.