Far Cry Review- NV SM3.0 vs ATI Sm2.0b [Archive]

]-[itman

07-26-04, 12:29 AM

Definitely a good read(for me anyway). According to that article, the x800's have the minimum number of instruction slots to qualify for SM3. If that's really the case, then it's pretty reasonable that given the right coding, the x800 cards would see a sizable performance increase equal or close to that of the 6800 cards.

I'm certainly no expert on the subject but I find the geometry instancing very interesting as well. Especially when they make this comment, "It is important to point out that ATI’s RADEON X800 XT and PRO graphics cards handle extreme geometry load better compared to NVIDIA’s GeForce 6800 Ultra and GT, which may mean that from this point ATI’s visual processing units have higher future-proof than NVIDIA’s latest graphics processing units do." I find this interesting considering the trend of the industry lately to more shader power and less pure geometry throughput, unless I'm missing something here? Definitely some interesting stuff here.

Steven4563

07-26-04, 12:55 AM

well in the screenshots i couldnt notice any difference what so ever

but it was a nice read before work :D

micamica1217

07-26-04, 03:04 AM

well in the screenshots i couldnt notice any difference what so ever

but it was a nice read before work :D

before....look. (http://www.xbitlabs.com/misc/picture/?src=/images/video/farcry20b/ati_pier_sm20b_bg.jpg&1=1)

after....LOOK. (http://www.xbitlabs.com/misc/picture/?src=/images/video/farcry20b/ati_pier_sm20b_veg100_bg.jpg&1=1)

I do like the ATI pics better then the nVidia ones in SM3.0, too many BLACK trees in the SM3.0 shots.....BLACKNESS. (http://www.xbitlabs.com/misc/picture/?src=/images/video/farcry20b/nvidia_pier_sm30_bg.jpg&1=1)

but that's just me... :-/

mica

scott_55X

07-26-04, 07:56 AM

This sums it up.
"Even though ATI considerably improved performance of the RADEON X800-series graphics cards in FarCry’s “Pure Mode”, NVIDIA’s GeForce 6800 Ultra still can claim performance leadership here, as in quite a lot of cases speed delivered by the GeForce 6800 Ultra is a bit higher compared to the rivalling RADEON X800 XT".

Sentential

07-26-04, 07:58 AM

before....look. (http://www.xbitlabs.com/misc/picture/?src=/images/video/farcry20b/ati_pier_sm20b_bg.jpg&1=1)

after....LOOK. (http://www.xbitlabs.com/misc/picture/?src=/images/video/farcry20b/ati_pier_sm20b_veg100_bg.jpg&1=1)

I do like the ATI pics better then the nVidia ones in SM3.0, too many BLACK trees in the SM3.0 shots.....BLACKNESS. (http://www.xbitlabs.com/misc/picture/?src=/images/video/farcry20b/nvidia_pier_sm30_bg.jpg&1=1)

but that's just me... :-/

mica

Thats PS3 at work. If you look closely, that is the shadow cast from the clouds onto the trees <not kidding>

bobmanfoo

07-26-04, 09:24 AM

what i gathered from this review, the x800xt and 6800u are pretty close, with the x800xt beating the 6800u in the highly intensive levels. it again shows that the GT and Ultra are very close. the x800pro falls behind the 6800GT in most of the tests, with the minimum fps close in some of them. I will bet my left pinky that with AA and AF turned on, we will see the same results as we have always, the XT will murder the 6800U and GT, and the x800pro will win most and lose some to the 6800u and GT

Albuquerque

07-26-04, 09:32 AM

Thats PS3 at work. If you look closely, that is the shadow cast from the clouds onto the trees <not kidding>
That doesn't make any sense, there are "dark" trees right in amongst the "light" trees, as well as there being no shadow on any of the other foliage nor the ground. Why would the ground be at normal brightness but a tree be dark?

I don't recall cloud shadows being part of this patch, nor HDR...

coldfusion71

07-26-04, 11:24 AM

here is a review for the x800 pro against the gt with aa and af http://www.elite*******s.com/page.php?pageid=5838&head=1&comments=1 furhter cementing the x800pro as a waste of money. put b a s t a r d s in the tilte it seems that that link got caught in the profanity filter.

Steven4563

07-26-04, 11:45 AM

mica i still cant see any difference lol :Confused: i got to also agree toh the Nvidia Screenshot that mica should has nasty dark bits i was thinking about getting a 6800 with my A64 system after my holiday aswell :(

coldfusion71

07-26-04, 11:51 AM

there is not any difference between the 2 cards, certain people claim to see a diffrence but they are only trying to justify thier purchase.

Mr.Guvernment

07-26-04, 12:11 PM

^^^ no, certain people just have better eye sight and an eye for detail then you do.

Albuquerque

07-26-04, 12:21 PM

there is not any difference between the 2 cards, certain people claim to see a diffrence but they are only trying to justify thier purchase.
What are you talking about? Did you even look at the screenshots that were linked? There are these arbitrary trees scattered throughout the scene that look like they were spraypainted BLACK. It's not justification or rationalization, it's called OBSERVATION.

And in case anyone needs a dictionary to understand what observation means:

The act or faculty of observing.
The fact of being observed.
The act of noting and recording something, such as a phenomenon, with instruments.
The result or record of such notation: a meteorological observation.
A comment or remark. See Synonyms at comment.
An inference or a judgment that is acquired from or based on observing.

Just making sure I have all the bases covered for the flamebait reply :)

bobmanfoo

07-26-04, 01:02 PM

lol

Albuquerque

07-26-04, 01:36 PM

I'm almost wanting to get the fubared 1.2 patch from somewhere else, just to try out Geometry Instancing on my ATI X800XT. I don't like the look of the flat vegitation sprites in the distance; I already manually specify a much higher vegitation LOD ratio even though it hurts performance.

micamica1217

07-26-04, 02:16 PM

Thats PS3 at work. If you look closely, that is the shadow cast from the clouds onto the trees <not kidding>

your kidding me right?

this patch is flawed in many ways.
I'm not suprised that the 6800 cards even get a large boost in framerate when SM3.0 is NOT used.
yet I'm still suprised the the ATI cards got up to a 20% performance decline with this patch, pre DX9.0b.

http://img.hexus.net/v2/features/farcry12/images/x800.png

with ONLY outside benchmarks tested, I have even heared that beside the fact that the 6800 cards may not render light sources from time to time, but the shadows are far harsher then ATI x800's...more jaggie.
the X-bit links are down for me, yet I have more IQ pics to show more on shadows.

now why should we see benchmarks INSIDE ???
because THAT'S were we will see the true performance of the cards, without being CPU limited...

LOOK. (http://discuss.futuremark.com/forum/showflat.pl?Board=techdisplayadapters&Number=4121608)

Hopefully this will help clear up when peeps ask is my CPU limiting a 6800/X800. You notice that outside is very CPU limited and the fps hardly changes between the res. However inside, where alot of shaders are used, especially for the lighting, the GPU carrys alot of the FPS and there is a huge diff between the res.

some added reading, that shows more info on SM3.0 and farcry....look. (http://www.ocforums.com/showthread.php?t=309362)

*mica puts on flamesuit, as he knows he'll be attacked far more them last nights scrim...lol*

mica

micamica1217

07-26-04, 02:21 PM

there is not any difference between the 2 cards, certain people claim to see a diffrence but they are only trying to justify thier purchase.

I guess your now ineligable to judge if nVidia's brilinear filtering is just as good or not, as trilinear filtering.

mica

micamica1217

07-26-04, 02:46 PM

bobmanfoo

07-26-04, 03:00 PM

^^^^ wow that is horrible :(

coldfusion71

07-26-04, 11:46 PM

more on shadows.....

ATI
http://www.xbitlabs.com/images/video/farcry30/radeon_shadows.jpg

nVidia
http://www.xbitlabs.com/images/video/farcry30/geforce_shadows.jpg

yeah, no IQ differance... :rolleyes:
if that's nVidia's "ultra shadowII" at work, then I don't want it.

mica

mica your posts are pretty onesided witn no objectivity at all, you dont state that the guy doing the review said that the blocky shadow was a driver bug, but you conviently left that out as usual. no other games displays that type of error there is no iq difference between the 2 card, i have had bboth and most other people who have had both disagree with your conclusions .also you have the nerve to post something about shadows, when talking about ati. how about posting all the games where the shadows are missing like splinter cell thief and other games ,how about the flickering textures in games dont act like at i is immune to having bugs that present themselves in games.

Steven4563

07-27-04, 01:00 AM

wow Mica they are really bad shadows i hope thats not what i missed when looking at the review or i must be blind LOL :eek: :(

micamica1217

07-27-04, 06:35 AM

mica your posts are pretty onesided witn no objectivity at all, you dont state that the guy doing the review said that the blocky shadow was a driver bug, but you conviently left that out as usual. no other games displays that type of error there is no iq difference between the 2 card, i have had bboth and most other people who have had both disagree with your conclusions .also you have the nerve to post something about shadows, when talking about ati. how about posting all the games where the shadows are missing like splinter cell thief and other games ,how about the flickering textures in games dont act like at i is immune to having bugs that present themselves in games.

I have never denied that at times, ATI drivers fix 5 games, yet break another in doing so...

this shadow issue, and the missing light sources on the trees seem MORE like the missing light sources and shadow problems in early aquamark and halo benchmarking...
in other words, A CHEAT.

since we have no other drivers to test the nVidia cards in SM3.0, and the fact that the authors of the two SM3.0 articals don't know what they are talking about....I wouldn't trust what they say.

here is at least one quote....

We also see almost no differences between NVIDIA’s and ATI’s image quality, except the “shadow issue” we discussed previously and some other minor things we are working on at this moment. While there are talks that NVIDIA encourages game developers to use 16-bit arithmetic precision instead of 32-bit precision (or 24-bit precision, like ATI) in order to improve performance, but possibly degrade image quality, we see no issues with NVIDIA’s quality under brief investigation.

what? you missed the missing light sources IQ pic that I pointed out?
how lame, and missleading of you.

NVIDIA seems to have a driver bug that results in very rough shadows in the game. This is unlikely to affect performance seriously as shadows in FarCry hardly require any significant computing power, though, this is something negative we can tell about the new set of drivers.

yeah, ok...go tell us that shadows in doom3 or farcry will not hardly require any significant computing power.
who do you think your readers are?

we the members of this forum, seem to know much more then the X-Bit authors.

now, one more thing....your the only blind man here.
as everyone can clearly see the IQ differance(s) I've listed above.
(and untill "I KNOW" that the issiues are just a simple driver bug, then I'll wait to say something that I know to be factual.)

there is not any difference between the 2 cards, certain people claim to see a diffrence but they are only trying to justify thier purchase.

you seem to be talking more about yourself in this quote, and many of your replys that attack my findings.

mica

coldfusion71

07-27-04, 09:24 AM

ok they are bugs when they are ati, but when they are nvidia they are automatically some underhanded cheating scheme, damn why didnt i think of that. all nvidia drivers from 61.45 and up support ps 3.0, but of course you didnt know that spouting off again. the missing light source is probably just an anomoly you need to stop with this lame ass mccarthyism crap.you have no evidence to back up anything you say ,you are making half asses assertions and speculating within your bias.lol i dont have to justify my card i didnt buy a 12 pipe card with terrible performance as you did. your finds are just your opinion which in the grand scheme of things is not that important, you shade facts to whatever you want them to be. who else has called those things cheats except for the great mica? is the driver set cat 4.7 a cheat because it had a artifact in 3dmark 2003? they did recall the driver so it must be cheating.

bobmanfoo

07-27-04, 09:28 AM

OBLIVIONLORD

07-27-04, 09:37 AM

I think both pics are screwy. THe 2.0 shows it as if the sun was right over certain trees but not all the trees. The 3.0 looks like small shadowy patches here and there over the trees. They both look very unrealistic and remind me of N64 grafics with better textures.

coldfusion71

07-27-04, 09:49 AM

how can you call that a bug? doesn't look like a bug, looks to like a shortcut. bugs usually are missing and flashing objects, objects being rendered incorrectly. and why is this? :

if you understand this, you would know that going from 16bit to 24bit or 32bit is not a linear increase in output but exponentially. in short, its sorta like comparing a nintendo to a super nintendo. why is does NV encourage 16bit? when done correctly and efficiently 16bits can't hold chit to 24bit much less 32bit arithmetic precision or anything else for this matter

alot of games have bugs with shadows on ati ,they are not being rendered correctly it looks like a bug to me. i should go through the game and see if it does it with every beta set of drivers i have the 62.20 now. certain shaders only require 16 bits to run at an optimum level without any iq degredation, others require higher precision. you are just wasting power to use a certain precision all the time ,it seems to me it is smarter to go the variable route. if you are not losing any quality in rendering using the variable method you are fine. the only reson why people brind this up because ati claims they are running 24which doesnt help image quality at all.

bobmanfoo

07-27-04, 10:04 AM

this will be the last post i will ever direct towards you. you accuse others of what you clearly do.
1. alwayshow do you know its an "anomoly"? please , explain yourself. i don't think you have owned both cards because of your comments.
2. and how is the x800pro have terrible performance when it can beat the snot out of the GT and Ultra with AA and AF is on in Far Cry? seems like you are the one trying to justify why your 16pipe card can't slaughter a 12pipe, like it should.
3. show me a CURRENT game where the GT slaughters the Pro..... now who is making half assed assertions? actually not even half but full. and you speak of 16bit and 24bit arithmetic precision when it seems you don't even know the jist of it, which do you think is more future-proof? why do you think NV prefers 16bit, when ATi is running at 24? why doesn't ATi just adopt 16bit again? and why is it that people always think NV is cheating? hmmm...... READ (http://www.extremetech.com/article2/0,3973,1086025,00.asp)
ok they are bugs when they are ati, but when they are nvidia they are automatically some underhanded cheating scheme, damn why didnt i think of that. all nvidia drivers from 61.45 and up support ps 3.0, but of course you didnt know that spouting off again. the missing light source is probably just an anomoly you need to stop with this lame ass mccarthyism crap.you have no evidence to back up anything you say ,you are making half asses assertions and speculating within your bias.lol i dont have to justify my card i didnt buy a 12 pipe card with terrible performance as you did. your finds are just your opinion which in the grand scheme of things is not that important, you shade facts to whatever you want them to be. who else has called those things cheats except for the great mica? is the driver set cat 4.7 a cheat because it had a artifact in 3dmark 2003? they did recall the driver so it must be cheating.

coldfusion71

07-27-04, 10:13 AM

i am doing the same thing as mica to counterbalance his fanboyism. i will give you my 9700 pro 9800 pro and my x800 pro benches to prove that i had the cards. i am well aware of the old nvidia cheating but, i think most and like me live in the present. you have no evidence to to say it is a cheat ,so why should i assume it is? why shouldnt i assume like when i had ati and games had missing shadows, flickering textures and artifacts that is it a driver issue. if no credible source says it is a cheat and you continue to assume it is youa re between a rock and a hard place.lol i trulely dont care what card is more future proof, i buy a new card every refresh. the only one i didnt was the 9800 xt because i had a tec on my 9800 pro and was overclocked to well beyong xt speeds.

http://service.futuremark.com/compare?2k1=6681871

http://service.futuremark.com/compare?2k3=2668142

bobmanfoo

07-27-04, 10:30 AM

here is a review for the x800 pro against the gt with aa and af http://www.elite*******s.com/page.php?pageid=5838&head=1&comments=1 [furhter cementing the x80pro as a waste of money. put b a s t a r d s in the tilte it seems that that link got caught in the profanity filter.
there is not any difference between the 2 cards, certain people claim to see a diffrence but they are only trying to justify thier purchase.
:clap: seems like there are still people out there who buy cards based on benches :-/ notice my questions were ignored

johnmcc516

07-27-04, 10:40 AM

looks like the 6800 GT wins my vote over the x800 PRO despite the shadow problem.

bobmanfoo

07-27-04, 10:46 AM

looks like the 6800 GT wins my vote over the x800 PRO despite the shadow problem.
are you telling me you would pay 4-5 hundred to play games with AA ans AF off?

Mr.Guvernment

07-27-04, 11:15 AM

i dont have to justify my card i didnt buy a 12 pipe card with terrible performance as you did.

lol yes the x800PRO is a HORRIBLE card - made from LAST gen ATI core and it can still beat NVIDIA brand spanking new architecture in various tests..lol would of been a sad day had ATI started a new core as well!

the x800 is not a horrible card - how many of them have you owned? i own one and i love it! - especially since I like to actually have AA turned on :D

coldfusion71

07-27-04, 11:16 AM

i dont buy based on base on benches, i buy based on gameplay and my gt wins over my pro hands down. noone is buying the pro anymore except for the vivo, which i might give a try. it is positioned terrribly against the gt and it cant compete ,it is a waste of money for a next gen card. show me a recent review that has the pro winning at all, most new benches put the gt ahead of the pro in about everything .it doesnt beat it at anything it was alot closer with the old drivers now it is more in favor of the gt.

Sentential

07-27-04, 11:20 AM

]-[itman

07-27-04, 11:22 AM

As far as cheating goes, you can't automatically accuse Nvidia of trying to cheat just because they've done it in the past, if you do that, you have to automatically accuse ATI of cheating every time a little something comes up because they've been caught cheating in the past too (Quake/Quak for those who don't remember).

As far as Farcry goes, obviously the 1.2 "patch" has issues with both lines of cards and to me, really should be a non-issue until it's fixed and any and all arguments about performance or "optimizations" are useless. We don't know just what in each area is due to driver issues and what is due to the patch issues.

As far as Anandtech and Xbitlabs goes, some are being way to hard on them. Neither of these two sites is biased to Nvidia or ATI. These sites are under A LOT of pressure to get things in, get them out, and get the results posted in as little time as possible, so obviously mistakes will be made and if mistakes ARE made, they should be made known to the site and they will fix what needs to be fixed. Just like what Anandtech did, they made an honest mistake which anyone in their position could have done, they realised it, and they fixed it.

Xbitlabs is simply working with what they are given. They are given the hardware, the latest drivers, and the latest official patch from Crytek. They start testing and post their results, exactly what they should be doing. It's not their fault if at the end of their review or even after they're finished, Crytek recalls the patch. They have their results and they post them, they also tell people that Crytek has just recalled the patch, so take the results as you will. If the reader then takes the review as gospel and as pure fact of performance, that's the reader's fault, not the reviewer's. As far as results go, I don't see what Mica is saying about Xbitlabs have different results from everyone else. First off, it's a given that each review site is going to have a bit different results because of all the factors that go into a system besides the video cards. The whole rest of the system can change from site to site and the results will be affected. Now, the good review sites try and isolate the subject at hand as best they can, but there's only so much they can do with the hardware they have and results will vary. Second, their testing showed a decrease in performance at times, depending on the map, when moving from the 1.1 patch to the 1.2 patch. However, once the ps2.0b render path is used, you see a jump in performance from 1.1 to 1.2. Third, they stated within their review that due to time constraints, they would not include AA and AF results in the initial review, but would update the review with those results, which they now have done. At the same time, this review's focus, especially now with the withdraw of the 1.2 patch, is not so much on the pure performance of each card, but rather the gains that can be accomplished by moving to ps2.0b and 3.0 respectively and why.

Again, I think that this whole issue should be dropped or atleast put on hold until the 1.2 patch is re-issued and the "official" results can be seen.

coldfusion71

07-27-04, 11:25 AM

are you telling me you would pay 4-5 hundred to play games with AA ans AF off?

i also run all of my games at 4aa 16 af so i dont know where you get that from. it is an anomoly until proven otherwise you have nothing to back up your asumption that it is a cheat.once again atis driver do the same thing and it is not considered a cheat, so why would it be a cheat for nvidia? if you haven no evidence you are just speculating innocent until proven guilty. considering the clockspeed disparity 350vs475 i think the gt is doing a hell of a job and add to that they overclock fairly well.

Sentential

07-27-04, 11:35 AM

Shadows Part 2

http://img.photobucket.com/albums/v180/Sentential/shadow2.jpg
^^ The shadows may be rendering properly. Look at the jagged vegitation. I think the only difference is the strength of ht shadows. nV's is much darker while ATi's is alot softer.

Personally a freaking shadow isnt gonna make me return this card. No way, no how.

Mr.Guvernment

07-27-04, 11:38 AM

honestally guys. It dosnt matter. Either way they are both good cards. No need for flaming people's purchases. As to the shadows arguement. The patch v1.2 fixed alot of it.

Proof:

http://img.photobucket.com/albums/v180/Sentential/FarCry0000.jpg
^^ That pic is from *MY* 6800GT, notice how crisp the shadows are.

very crips - but your gun looks very jagged around the edges. especially around the sight and top handle - you using AA or AF or anything? (yes i clicked on the image for the full screen version)

As for the GT beating the x800PRO in all benchs - got links? do these links use AA and AF to the max? i am not tryin got defend my purchase but most reviews i still see today they break even.

Sentential

07-27-04, 11:40 AM

very crips - but your gun looks very jagged around the edges. especially around the sight and top handle - you using AA or AF or anything?

Nope thats with AA and AINSO off at 1200 / 1076. I never thought to kick AA or AINSO on again, since my FPS is so high and it looks great. However I am gonna do that now, and re-post a pic

bobmanfoo

07-27-04, 11:53 AM

Listen, i was referring to this review. you need to stop making assumptions. just ignore me like i will you since you can't seem to hold a conversation.
i also run all of my games at 4aa 16 af so i dont know where you get that from. it is an anomoly until proven otherwise you have nothing to back up your asumption that it is a cheat.once again atis driver do the same thing and it is not considered a cheat, so why would it be a cheat for nvidia? if you haven no evidence you are just speculating innocent until proven guilty. considering the clockspeed disparity 350vs475 i think the gt is doing a hell of a job and add to that they overclock fairly well.

coldfusion71

07-27-04, 11:54 AM

if you post something erroneous, i will call you on it. i dont really care if you ignore me.

Mr.Guvernment

07-27-04, 11:56 AM

here is a pick i took with my x800PRO in the gam Joint Operations Typooh Rising - NOW i did use the ingame screen shot button so not sure what quality they save this @ - this pic is @ 1280 x 1024 4xaa and 16x AF

(i amnot sure how good nova world coding is compared to crytek so not sure if that would matter) - i will take some better shots tonight - this was just a pic i grabbed because of the 2 guys glitching which i am reporting to nova) the ont thing i noticed is the different level or darkness in the shadows under the guys - to me that is detail.
be sure to click it to get full screen - as said will try to take better SS tonight when i play :D
http://66.207.201.54/matt/SS00001.jpg

Sentential

07-27-04, 11:56 AM

OK... here is a shot with AINSO and AA on full.

http://img.photobucket.com/albums/v180/Sentential/FarCry00034.jpg
No jaggies, and I see 0 difference between it and my 9800PRO

bobmanfoo

07-27-04, 11:58 AM

if you post something erroneous, i will call you on it. i dont really care if you ignore me.
how is what i said erroneous? DID THIS REVIEW USE AA AND AF? the person i was responding to was concluding this review. you'll call me on it.... please. i called you out on alot of your posts, and what did you have to say? NOTHING and you still haven't answer my questions.

Mr.Guvernment

07-27-04, 11:59 AM

Nope thats with AA and AINSO off at 1200 / 1076. I never thought to kick AA or AINSO on again, since my FPS is so high and it looks great. However I am gonna do that now, and re-post a pic

Sounds great - tonight i will get some good pics with AA and AF off as well on this card @ 1280 x 1024 - would be nice to see screenshots from games that are not on the top 10 benchmarkers list - this way it is less likely either company has any optimizations for that game :)

On that note i loe my x800PRO - it is big difference from my 9700PRO - if i can financially - i would love to get a 6800GT or ultra in the near future just so i can have mny own fun and comparisions - and so i can have 2nd kick arse rig that my friends can use to plan lan game son :)

Mr.Guvernment

07-27-04, 12:00 PM

Sentential - that is a dam nice pick i must admit - and i know the jaggies on his glove is the actual graphics they used for the glove :) very crisp and clean.

Sentential

07-27-04, 12:01 PM

So *hope* this settles things. It is software issue. And *NOT* the hardware itself. I mean hell. Do we really have a right to complain about software bugs on brand new hardware??

Sentential

07-27-04, 12:02 PM

Sounds great - tonight i will get some good pics with AA and AF off as well on this card @ 1280 x 1024 - would be nice to see screenshots from games that are not on the top 10 benchmarkers list - this way it is less likely either company has any optimizations for that game :)

On that note i loe my x800PRO - it is big difference from my 9700PRO - if i can financially - i would love to get a 6800GT or ultra in the near future just so i can have mny own fun and comparisions - and so i can have 2nd kick arse rig that my friends can use to plan lan game son :)

D00d I love the x800 series too. I have no bias whatsoever. I was gonna buy the first thing I found. If I could get an x800 before I got my 6800GT, I would have one now.

I do agree. These cards are at a whole new level. I cannot praise them enough. :burn:

coldfusion71

07-27-04, 12:03 PM

So *hope* this settles things. It is software issue. And *NOT* the hardware itself. I mean hell. Do we really have a right to complain about software bugs on brand new hardware??

hey sentential what are you using to capture your pic i need to take some screen shot fo r a buddy of mine.

Sentential

07-27-04, 12:07 PM

"F12" in the game takes a screenshot. :attn:

coldfusion71

07-27-04, 12:11 PM

sweet thanks man

Mr.Guvernment

07-27-04, 12:12 PM

^^^ i am going to find some 3rd party stuff - the in game screens tend to use alot of jpeg compression :(

bobmanfoo

07-27-04, 01:16 PM

heres NFSU at 1280x1024 AAx4 AFx16 IQ set to High Quality, 61.76WHQL
PIC (http://www.msnusers.com/myrig/shoebox.msnw?action=ShowPhoto&PhotoID=56)

Mr.Guvernment

07-27-04, 01:54 PM

i will install that game tonight if i dont fall asleep and get some screens - u just use f12 for it as well?

time to break out the steering wheel! - even on my 9700POR i could runt he game full details @ 1600 x 1200 - no aa or af though i dont think.

What car is that?

Sentential

07-27-04, 01:57 PM

As I said b4. The 6800GT is great. There are no real image problems, they are for the most part..... bugs.

micamica1217

07-27-04, 02:51 PM

As I said b4. The 6800GT is great. There are no real image problems, they are for the most part..... bugs.

I have to agree with you sentential to some degree.

if you reread my original reply to someone, I said, they seemed like the performance inhancing, missing light sources, like in halo and aquamark.
if the missing light sources are just unintentional bugs, that's more then fine with me.
yet it does look like what happened before with nVidia.

cheats or not, not rendering light sorces will give you a performance increase in benchmarks...
yet for ]-[itman, I don't remember talking too much about benchmark numbers.
I was talking about IQ and how the autors don't realy know what they are talking about.

Sentential,

thank you for your original pic on the shadows of the gun....
could you try something for me?

in farcry, it seems that there are TWO types of shadows from trees.
one is LIGHT, like on your pic...
the other is DARK, like in the examples from the x-bit site.

please try with AA and AF turned on...a new pic with the DARK shadows.
just so we can still see how "see through" and jaggy they are, or are not.

-----
I'll also testify, that you could have gotten an x800pro, as I've PMed with you on the very same matter.
-----

I didn't mean to start a flamewar or cause any problems with members here.
I posted some findings, and it seems that some people can't handle anything that doesn't show nVidia in a good light.

if anyone still thinks I'm ATI biased in any way...I"M THE ONE WHO POSTED THE ATI RECALL CAT4.7 THREAD!!!!!

and to anyone reading this....you are not to flame anyones' purchace of any hardware.
doing so, could get you a vacation.
also calling what I posted, the IQ pics, quotes and so on, in a negative light...
is a flame against me. (that's a no, no.)
some people are walking on some very thin ice...I'm not a mod, but you are warned in advance.

mica

Sentential

07-27-04, 03:08 PM

http://img.photobucket.com/albums/v180/Sentential/0.jpg

Here is a shadow from a building. There is a little bit of "swimming" but its not nearly as bad as the pic you showed me. Lets see some of your pics. I wanna see the difference.

micamica1217

07-27-04, 03:22 PM

Here is a shadow from a building. There is a little bit of "swimming" but its not nearly as bad as the pic you showed me. Lets see some of your pics. I wanna see the difference.

I don't have hosting at this time...sorry, no pics. :cry:

if your running with DX9.0c, the same driver from the review, and the patch1.2, SM3.0....then I'll say that shadow is nice/fine/no error, yet still not from a tree.
we need a pic that in many ways, replicates, the pics I grabed from x-bit, and with the same type of variables.;)

in farcry, it seems that there are TWO types of shadows from trees.

we realy need a pic from under a tree.

one is LIGHT, like on your pic...
the other is DARK, like in the examples from the x-bit site.

thanks in advance for your hard work.

mica

bobmanfoo

07-27-04, 03:26 PM

1280x1024 is the max res., i had to use the prt Sc button, or maybe windows button + prt sc button, not sure, i hit both, i think that is a miata
i will install that game tonight if i dont fall asleep and get some screens - u just use f12 for it as well?

time to break out the steering wheel! - even on my 9700POR i could runt he game full details @ 1600 x 1200 - no aa or af though i dont think.

What car is that?

JoT

07-27-04, 03:33 PM

Hey guys, I'm thread watching. Please play nice :)
Just a friendly reminder from your neighborhood Triforce wielder.

micamica1217

07-27-04, 04:02 PM

JoT,

OMG....ROTFLMAO

how did I miss what's in your siggy?

-"Man, McGuyver could defuse a thermonuclear device w/ a paperclip and a stick of gum." -- Oni
-"If it ain't watercooled then it should hover with fan power" -- Phantom
-"actually id go with mushkin it is quite possibly the best most relaible ram ever made, in the history of forever." --Robrules18
-"I did not become a vegetarian because I like animals. Rather it was because I hate plants." --Malpine Walis
-"I could teach you to fly a plane, but from my experiance with Flight Sims, it would be a one way trip, resulting in a crater." --9mmCensor
-"oh hello there Mr. Spider! You are the size of a cat arent you.....whoops what's this.....it appears my arm is missing....hmmm..." -- @md0Cer

keeping it light,
mica

Steven4563

07-27-04, 04:38 PM

for people with NFS:U i can run that game on my comp in sig with 1280*1024 4xAA and 8xAF.

i have to agree tho with JoT play nice, ive had my warnings and my final warning by someone else why u think im keeping out off this one :D

also them farcry pics are very nice :attn: apart from Sententials inside pic i dont like the texture of the gun i dont know weather its the game or not but its kinda looks blurred to me

Mr.Guvernment

07-27-04, 04:52 PM

LOl
DUH me for some reason ithought it could do 1600 x 1200!!!

Any test anyone want me to run - i would try farcry but it seems the ONLY gaem that freezes on me with ym card overclocked to 530 / 560 even - yet ALL my other games run fine

Joint Ops
battle field 1942
MOH - all add-ons
Call of Duty

:(

dam u farcry!

i will take purdy shadow pic's on Joint Ops just so i can feel special :)

]-[itman

07-27-04, 05:06 PM

yet for ]-[itman, I don't remember talking too much about benchmark numbers.
I was talking about IQ and how the autors don't realy know what they are talking about.

My mistake, I was confusing this thread with another, my apologies. However, I still think that you are being far too hard on xbitlabs. Their time spent with the new patch was brief and they make a point to say that their analysis thus far was only a brief examination of the patch. From what I've read, what xbitlabs has said is just about what I've read from Crytek reps(not just pr guys). Their analysis of the patch seems to be pretty informative to me, even if it was rushed a bit and there are a few minor errors.

hUMANbEATbOX

07-27-04, 06:53 PM

heres NFSU at 1280x1024 AAx4 AFx16 IQ set to High Quality, 61.76WHQL
PIC (http://www.msnusers.com/myrig/shoebox.msnw?action=ShowPhoto&PhotoID=56)

:eek: what's with the horrible jaggies on the roadside curb??

and around around the back window? or is that just the cars rearwindow texture that sucks?

theELVISCERATOR

07-27-04, 07:25 PM

just throwing in an x800 screenie for comparo...

http://home.comcast.net/~eknop/FarCry0012.jpg

render mode 3 rocks huh? I think it looks far more photo realistic myself...

Mr.Guvernment

07-27-04, 07:57 PM

I dont worry about [H] Doom3 benchies, They were cpu limited!

LOL thas great - and nice pic :)

micamica1217

07-27-04, 08:34 PM

just throwing in an x800 screenie for comparo...

http://home.comcast.net/~eknop/FarCry0012.jpg

render mode 3 rocks huh? I think it looks far more photo realistic myself...

great example of what we need to see....same gun, same dark shadow.

great work

mica

bobmanfoo

07-27-04, 10:04 PM

:eek: what's with the horrible jaggies on the roadside curb??

and around around the back window? or is that just the cars rearwindow texture that sucks?
thats why i posted the pic, its not the game or the texture, its either the card or the driver, dunno what is up with the chitty AA

micamica1217

07-27-04, 10:55 PM

thats why i posted the pic, its not the game or the texture, its either the card or the driver, dunno what is up with the chitty AA

does the game have AA in the UI???
if it does, try turning off AA in the CP, like you have to do with farcry.
retest.

mica

bobmanfoo

07-27-04, 10:57 PM

does the game have AA in the UI???
if it does, try turning off AA in the CP, like you have to do with farcry.
retest.

mica
nope, i can only do it through the CP

Sentential

07-27-04, 11:13 PM

http://img.photobucket.com/albums/v180/Sentential/FarCry0010.jpg

Sorry guys, I dont know how to enable cheats to get that gun up. This is the best I can do. I can see the pixelation, but it seems like a bug to me.

Sentential

07-27-04, 11:30 PM

No bias, no agenda. Here is a pic.

http://img.photobucket.com/albums/v180/Sentential/FarCry0005.jpg

^^ The jaggies are present. I still stand by what I said. Its a bug. Im using DX9c, v1.2 and all that good stuff. I hope this is addressed soon.

Mica, despite what you say about Ultrashadow II, *that* is what gives the nVidia the performance boost. If you dont belive me, look at 3dmark03.

Notice how nVidia is +20FPS ahead of all the x800s on "The Battle of Proxycon", and how it blows in nature. *THAT* is Ultrashadow in action.

micamica1217

07-27-04, 11:39 PM

Mica, despite what you say about Ultrashadow II, *that* is what gives the nVidia the performance boost. If you dont belive me, look at 3dmark03.

Notice how nVidia is +20FPS ahead of all the x800s on "The Battle of Proxycon", and how it blows in nature. *THAT* is Ultrashadow in action.

hold on there bro....

I've never said that "ultra shadowII is not a great performance booster.
and I agree that it's most likely a bug, that will be fixed.

once again, thank you for your hard work.

mica

Sentential

07-27-04, 11:42 PM

hold on there bro....

I've never said that "ultra shadowII is not a great performance booster.
and I agree that it's most likely a bug, that will be fixed.

once again, thank you for your hard work.

mica
Heh, my bad. I only saw the sarcastic comment about it :p. But yes, even you can agree that it jacks the FPS up alot, especially in Doom3.

As I said b4. I dont have a bias at all. I just wanna level the playing field. nV continues to have IQ problems. <big surpise :rolleyes:> But still the performance is there. Something which the FX series lacked horribly

Sentential

07-27-04, 11:45 PM

One more thing I would like to add. Is it me or does my screenie look alot more "chromatic" / brighter / more vibrant than the x800 shot.

Im not knocking it, but mine looks more like a painting and not peticularly photographic, while the x800 looks bleeched and darker. Giving the effect of a photo.

Its very odd to look at another rendering device. nV's rendering path feels more "surreal" than does ATi's or maybe its just me.

theELVISCERATOR

07-28-04, 12:05 AM

snyper1982

07-28-04, 12:08 AM

i prefer the look of the ATI personaly. That being said, IF I pick up a new card this round, it will most definately be a 6800GT. Price performance favors it, and thats what im all about :)

Mr.Guvernment

07-28-04, 01:06 AM

^^ the GTs i am seeing aroud $400 - $450 on newegg - the x800 are also $400 - $450

snyper1982

07-28-04, 02:19 AM

exactly, from what i have seen the GT has slightly better performance. but i doubt very serioulsy i will be getting a card this round. most likely next gen.

verbatim

07-28-04, 06:56 AM

I would never have noticed the differance ib game looking at screens like those two. They both look great. Only when flipping between them can I notice the differance in the trees. Is it there? yes Would I have ever missed it? No.

germanjulian

07-28-04, 07:32 AM

FLAMER BABIES!

my graphics card is better then yours :attn:

if you guys would acctually read the conclusion and switch on your brain instead of using your Nvidiot and FanATIc brain cells you would realize that...

why do I bother... just read the conclusion.

Sentential

07-28-04, 08:54 AM

its the render mode... cold....looks way better in game

try this at console...
r_rendermode 3
Thnx for the heads, "Cold" does look infinatly better than anything else :beer:.

Some pplz take this whole ATi and nV thing a bit too far. Especially now since they are so close. Personally I think its a matter of preference and situation.

In my case a 6800GT made sense, just like an x800s did for you and mica. Personally you cant go wrong with either. They both have their strengths and weaknesses

x800PRO

PROs: very efficient AA / AINSO rendering. 3DC. Has a much higher minimum FPS
CONs: Inefficient GPU, Price, availabilty, 12 pipes.

6800GT
PROs: Higher raw FPS. VS3.0, PS3.0. Ultrashadow II <--One of the best besides SM / VS 3.0. Extremely efficient GPU.
CONs: That god damned thermal throtteling, and how it gets worse with AA / AINSO....basically Heat. :temper:. Poor efficiency with AA and AINSO, although its better now. Lower IQ.

Mainly here are the big problems I have with the x800. #1 *Horribly inefficient GPU* and that it dosnt *fully* support DX9c.

Main problems I have with the 6800 is ther *FRIGGING* thermal-throttling, which will be fixed soon. No GPU should ever go to 78c* after 1 lousy pass of 3d03. Thats just ridiclous.

micamica1217

07-28-04, 10:06 AM

Thnx for the heads, "Cold" does look infinatly better than anything else :beer:.

Some pplz take this whole ATi and nV thing a bit too far. Especially now since they are so close. Personally I think its a matter of preference and situation.

In my case a 6800GT made sense, just like an x800s did for you and mica. Personally you cant go wrong with either. They both have their strengths and weaknesses

x800PRO

PROs: very efficient AA / AINSO rendering. 3DC. Has a much higher minimum FPS
CONs: Inefficient GPUhow so? please explain to me, how this "inefficient" GPU can sometimes beat a 6800u and still be called an inefficient GPU?, Pricegoing for about the same price, I don't see your point., availabiltyI got mine before the 6800gt/u were even out on the market yet...if I can get this card, then anyone can., 12 pipes.

6800GT
PROs: Higher raw FPS. VS3.0, PS3.0. Ultrashadow II <--One of the best besides SM / VS 3.0. Extremely efficient GPU.
CONs: That god damned thermal throttelingPM me, and stop crying about the heat. I might have a fix for your throttleing...it requires a flash of the bios that increases your cards voltage by .1v , and how it gets worse with AA / AINSO....basically Heat. :temper:. Poor efficiency with AA and AINSO, although its better now. Lower IQ.

Mainly here are the big problems I have with the x800. #1 *Horribly inefficient GPU* and that it dosnt *fully* support DX9c.

seems to me that since you got this 6800gt, your starting to listen to the fanboyish stuff on other forums....
unless your really joking when you say *Horribly inefficient GPU*

Main problems I have with the 6800 is ther *FRIGGING* thermal-throttling, which will be fixed soon. No GPU should ever go to 78c* after 1 lousy pass of 3d03. Thats just ridiclous.

*wondering what's going on with my fellow OCer?*

mica

]-[itman

07-28-04, 10:27 AM

I agree with Mica, how is the x800Pro at all ineffecient when it being a 12 pipe card, can stand up to the 6800GT, a 16 pipe card, and sometimes beat it(when AA and AF are cranked up). To me, that's a very efficient core, not ineffecient. Maybe you're just going by the clock speed alone? Well, there's a lot more to a core being efficient or not than that. You have to look at capabilities of the card, heat output, transistor count, as well as other things to see if the chip is really effecient or not. There's also the factor of software(drivers, different games work better with different chips) you have to look at. I just don't see how it could be an ineffecient card.