x2 vs conroe/core2duo gaming, the real story [Archive]

AshenSugar

10-09-2006, 01:03 AM

http://enthusiast.hardocp.com/article.html?art=MTEwOCwxLCxoZW50aHVzaWFzdA==

The Bottom Line

We have proven here that the flurry of canned benchmarks based on timedemos showing huge gains with Core 2 processors are virtually worthless in rating the true gaming performance of these processors today. The fact of the matter is that real-world gaming performance today greatly lies at the feet of your video card. Almost none of today’s games are performance limited by your CPU. Maybe that will change, but given the trends, it is not likely. You simply do not need a $1000 CPU to get great gaming performance as we proved months ago in our CPU Scaling article.

When it comes to playing games, the only persons that need to be even a little concerned with upgrading their CPU to a Core 2 processor might be those with high-end SLI, CrossFire, or GeForce GX2 video cards and we have yet to prove that due to the testing limitations we ran into. Then, and only then, you might see an Intel Core 2 processor deliver a performance advantage.

Lastly, I would advise everyone that is thinking of rushing out and purchasing their latest upgrade that we are sure to see HUGE pricing slashes out of AMD before the end of the month.

read the artical fully to understand, but only the x6800 and e6700 beet the fx62 and then it was marginal.

for a gamer its better to get the best gfx card u can, then 2gb decent ram, then get board and cpu as your last step going for the best bang for the buck cpu/mobo you can get, i know some would say that "when new faster cards come out conroe/core2 will gain more" but by then you will also betable to get k8l/barcolna and other newer amd cores that will have been perf enhanced, and i kinda dought from the results with SLI 7900gtx and x1950xtx cards that u will see much of a diffrance between core2 and x2

cant remmber where the bench was, but it was a small diffrance, 2x x1950xtx was fastest solution but whos gonna spend that kinda money on video cards when u can just save ur $ and buy a new next gen card when they come out and get all the new fetures and perf boosts said cards bring :)

10-09-2006, 01:23 AM

i said it before and i say it again
as long it is above 35fps you won't see the diference

if you want to crossfire a couple x1900xt you are just a benchmark whore, nothing more
as for cpu's i stick with my 3000+@2.7ghz that is pretty good :)

Azn Tr14dZ

10-09-2006, 02:00 AM

10-09-2006, 02:04 AM

2x X19x0's in CF might be nice if you play in high resolutions and play at max settings and demanding games. But it's useless to go CrosFire or SLI when you only play at low resolutions...I see some kids getting 2 X1900XTX's and going CrossFire. And then I ask them "what resolution do you play at?"...and they say "1280x1024 w/ 2x AA and 4x AF"...what a waste...

just :nutkick:the kids and teach them they are spending too much money, crossfire is in my opinion useless, a single x1900xtx can play all games at max settings above 35fps :shadedshu

Alec§taar

10-13-2006, 04:21 AM

No Quake 4 1.3 SMP multithreaded gaming on this test, a test of dualcore CPU's?

:(

* The test's not comprehensive enough!

I say this, simply because of this - it's NOT leveraging the extra cores fully if none of those games bear more than 1 thread & are not thus, 'implicitly ready' for SMP computing on modern OS' then...

(& certainly not, if they do not use manual 'explicit multithreading' via actual processor #'s detections + cpu usage on each CPU, & SetThreadAffinityMask Win32 API calls, etc., to do so).

APK

P.S.=> Are ANY of those games they tested w/ multithread designed @ least, if not explicity coded for SMP as noted above?

YES: Their test, it is "head-to-head",... but, w/ ALL SINGLE THREADED GAMES!

(Yes, single-threaded games should see some small increases too, MAINLY because the OS can send other running processes to less used CPU's giving MORE on 1 of the N cpu's to the game iself, & possibly only THE GAME itself)...

However, imo? This test's incomplete!

Again: As it's not using what dualcore CPU's are for in SMP setups, testing single thread designed games only, omitting multithreaded games like Quake 4 SMP 1.3 (which I know, with certainty, IS multithreaded) in its battery of tested games, fully exploiting what that type of design gives you on an SMP setup! apk

15th Warlock

10-13-2006, 04:39 AM

However, imo? This test's incomplete, as it's is not using what dualcore CPU's are for in SMP setups, testing single thread designed games only, omitting multithreaded games like Quake 4 SMP 1.3 in its battery of tested games! apk

And no benchmarks of COD2, FarCry, City of Heroes, City of Villains, or Serious Sam 2 neither, all well known dual core enhanced games, suspicious, huh? :p

EDIT: Not so surprising after seeing who wrote the article though...

AshenSugar

10-13-2006, 04:47 AM

this wasnt an smp test, that talks in another thred, this was simply showing that core2 isnt really any better for gamers who dont got $1200+ video card rigs, cf/sli would show benifit, but currently i know only 3 people who have sli or crossfire, one has both, x1900xt cf and 7800gtx sli, one has 6800gt sli, one has 7600sli and a x1800gt sli, the last 2 are not in the running for this kinda test because they are older/lower end cards, the first one has them in dual dual core optiron setups(for work, hes a game designer, needs to beable to test on nvidia and ati hardware) almost forgot the guy with the x1900xt and 7800gtx rigs also has an orignal x800xtpe crossfire rig, that was his game rig for home, its now used for sm3b path testing and for his little bro to game on when he comes to visit(few times a week)

the point is that it shows the common games people play with fairly common hardware for a "true pc gamer" i would have liked to have seen it with x1900xt card as well just to see the diffrance's.

quake4/doom3/prey are SMP enabled, and as i understand it fallback to a diffrent mode when running on singel core chips so that perf loss isnt drastic tho it is detectable, you can run multi threded apps on singel core systems, the threds just have to time share on the cpu.

both amd and intel are shown fairly here from what i can see, sure they didnt bring in spicific games to test both cores at once, but thats gonna endup with the same result, core2 will gain from its extra power but could easly loose some perf(2mb cache versions may feel pain if to much data is moving thru the fsb tho) things for singel card gaming at reasonable res's are pretty even right now, no need for a gamer to rush out and grab core2 when for the price of an e6300 chip they could get a 3800+ x2+overclocking board and get the same results with there video card.

google around i use to have links to some benches with smp enabled games on dual core vs singel core, p4 got a decent boost when it had HT enabled, but still got soundly stomped by amd's offerings :)

AshenSugar

10-13-2006, 04:56 AM

And no benchmarks of COD2, FarCry, City of Heroes, City of Villains, or Serious Sam 2 neither, all well known dual core enhanced games, suspicious, huh? :p

EDIT: Not so surprising after seeing who wrote the article though...

farcry not dual core/smp
serious sam 2 not dual core/smp
city of heros isnt benched its an mmo same with cov

cod2 isnt commonly benched by many sites as its not as popular as other games such as quake4/prey/doom3 and halflife2

the test they did was on some of the most commonly played games, they had to choose a list of games to bench to keep it resonable, personaly i understand why many sites now limmit the number of games they benchmark, benching takes alot of time as you normaly need to run each bench at least 3 times to get a valid number, also the caust of buying the games gets a bit up there, 40bucks a pop minimum for newer games restricts alot of bench sites from buying every new game out there, demos arent a good example because most demos dont get the same code as full version and def dont get the updates put into the full versions.

i use to work for a guy who ran a hardware/tech site, testing gets expencive, you pay the person doing the tests as well as buying all the testing software, you cant use warez to benchmark pc's afterall :P

in short, yes it was a short test/bench, but it was only done to put things in prespective after intel hyped the results of conroe being 10000000x better then amd for gaming, sure if you have quadsli, high end sli, high end crossfire you will see a diffrance in fps but most people dont wana spend that kinda $ on a video card, hell i had a problem paying 370for my x1900xt(270 after MIR) at the time, and i could have afforded a quad sli or x1950xtx crossfire, but i dont see the point, i would rather spend my $ on other things or save it to buy a 2nd gen dx10 card when they acctualy start being used(first ge alwase endup being trumped pretty fast afterall)

Alec§taar

10-13-2006, 05:46 AM

IMO, & it seems those of others?

This test is incomplete!

The graphics card has more weight, but they're not showing what a game like this can increase by via using SMP ready code & on an SMP rig...

& Quake 4 1.3 SMP's shown huge increases in the ranges & conditions especially which I noted below in my P.S., via this CPU driven gain... PURE CPU DRIVEN GAIN, mind you, not graphics card driven gain only.

This test TOTALLY omitted multithreaded code in that test, but that would tend to make what they said ("videocard = everything") in this test, on SMP rigs? VERY suspicious!

(& they told of lies & such?):

The fact of the matter is that real-world gaming performance today greatly lies at the feet of your video card. Almost none of today’s games are performance limited by your CPU

That's NOT the TRUTH @ ALL, & Quake 4 SMP beta's tests showed it in BETA form no less (slower than final code)!

It's not true... & we all have seen proof here before in fact... in a real world test with THIS game on SMP rigs, that it's not & mind you, from reputable sites!

AMDZone being one, & I put up others here before also, showing up to 87% CPU Driven Gains, averaging around 45%!

(& Quake 4's a realworld test, w/ a real SMP ready game, not synthetic)...

APK

P.S.=> Their omitting testing multithreaded game code, by NOT doing a test of SMP ready game code (E.G.-> Quake 4 1.3 SMP build)? Especially on DualCore machines from AMD & INTEL??

* Glaring omission in this type of test...

Especially @ resolutions from 640x480 - 1024x768 ranges, where the CPU being dualconfig has been shown to help the game perform better in, rather excessively...

(Less w/ AA & AF turned on fully, but still there showing SMP configs help that game gain speed, & all the way (albeit decreasing) into the highest possible resolutions)... apk

AshenSugar

10-13-2006, 07:29 AM

who would buy a x1900 or 7800/7900 to play at 640x480 or 1024x768?

that was there point, nobody whos sain is gonna spend 300bucks on a video card to play at those res's even with aa/af maxed that isnt even touching the video card, people who play at thoses reses will beusing 7600 or lower cards and thus again they will be gpu limited, sure a PURE cpu test at low res shows conroe/core2 killing amd, but common, what self respecting gamers gonna go back to low res's on a gaming card?
only 1 group i can think of would do that, CS kiddies, the kids who turn everything down and run as low a res as possable to get 300+fps because they dont get that 60 range is PLENTY for shooter games.

email them if you want them to redo the benches using smp patched quake4, maby they will ablige you, personaly i feel that if you read what they where doing you would understand that it was to show what real common gamers would run, not what cpu bound resolutions will give you.

say they go SMP enabled and its balancing between the cores, the video card is still gonna be the limiting factor, not the cpu's/cores, the video card even if it was a x1950xtx is gonna be the bottleneck at any reasonable gaming resolution(reasonable for the card in use)

intels contention was that when running core2 you would see huge gains in gaming, they didnt mention that there tests where on intel tweaked CrossFire and SLI rigs, or the how the tests where done, many of the tests where run at 1024 res(my buddy works for there portland test labs) they tested/benched using what would make them look good, not what the avrage pc gamers gonna have OR use for gaming, hardware wise and game setting wise intels numbers are unrealistic, as unrealistic as you seem to think the xbit review/artical is because it leaves out XXX game or XXX patch.

cjoyce1980

10-13-2006, 08:14 AM

i have my 6800LE and a Athlon XP 3200+ and i can run all my games max out a 800x600. this all comes down to personal choice and the end of the day.

AshenSugar

10-13-2006, 11:21 AM

wouldnt reccomend that low a res with an lcd, they are made to run at 1 default res, changing res messes with quility, corse anybody who games at 800x600 probbly wouldnt know the diffrance.

and a 3200+ and 6800LE(eww) would keep me from going above 1024x768 anyway, the gfx card is the bottleneck in that system.

whats with the lanparty rig being better then ur main rig tho :P

Alec§taar

10-13-2006, 01:21 PM

who would buy a x1900 or 7800/7900 to play at 640x480 or 1024x768?

The SMP based code (CPU Driven gain) gain is STILL THERE @ that resolution as well (not as large as it is in lower resolutions, but there nonetheless)... a CPU-driven gain.

Omitting it makes their statement of:

The fact of the matter is that real-world gaming performance today greatly lies at the feet of your video card. Almost none of today’s games are performance limited by your CPU

Seem TRUTHFUL, and it is NOT.

that was there point, nobody whos sain is gonna spend 300bucks on a video card to play at those res's even with aa/af maxed that isnt even touching the video card

Why on earth would anyone use FULL AA/AF @ resolutions above 800x600?

Do the folks who wrote that article even understand WHY AA/AF are used??

It is for resolutions where "jaggies" (image edge's distortions) or lack of image clarity/definition are present & happening... which is NOT @ HIGHER RESOLUTIONS NEARLY AS MUCH (especially regarding jaggy edged imagery)!

Hence, why even use them @ 1024x768 or more, if those resolutions DO NOT DISTORT IMAGES MUCH, IF @ ALL?

email them if you want them to redo the benches using smp patched quake4, maby they will ablige you, personaly i feel that if you read what they where doing you would understand that it was to show what real common gamers would run, not what cpu bound resolutions will give you.

Do these "real gamers" understand WHEN to use AA/AF, & what they do?

Using AA/AF @ res' over 800x600? A WASTE of power & unnecessary to perform (especially in the case of AA @ high res gaming)!

At HIGH RESOLUTIONS, gains were shown @ AMDZone:

http://www.amdzone.com/modules.php?op=modload&name=Sections&file=index&req=viewarticle&artid=225&page=2

ATI x1800XL videocard Quake 4 SMP test results (showing CPU-driven gain):

http://www.amdzone.com/pics/gaming/quake4/dualcore/x1800xl.jpg

NVidia GeForce 7800GT videocard Quake 4 SMP test results (showing CPU-driven gain):

http://www.amdzone.com/pics/gaming/quake4/dualcore/7800gt.jpg

"their SMP patch does wonders with 50% improvements and is a must download for every dual core owner out there, AMD or Intel. If this isn't enough we also have word from Tim that in the 1.1 update to Quake 4 we will see even more SMP improvements and fixes, so we will definitely be testing that out. Also where id leads, others follow. Obviously for id to patch the game and enable this SMP support and show what a huge advantage it can have goes against some of the naysayers earlier. Hopefully we start to see more companies utilize this second core in coming time and congrats to id software for being the first."

& other tests done @ websites concerning SMP ready code in Quake 4 SMP showed cpu-driven gains, even @ highest res' they tested in WITH AA/AF on FULL!

* CPU-driven gains via the use of SMP-ready code in Quake 4 SMP betas (which won't show as much as full builds would, due to beta debug symbols & code still present in the game engine), not GPU gains only.

:)

APK

P.S.=> I feel that due to them omitting this data? They are 'short-changing' folks, saying "GPU = everything" & "CPU = nothing" basically when it comes to gaming performance... perhaps in single-threaded games, yes... only small gains result that are CPU-driven in that case!

(As I stated last page w/ detail on why or how: I.E.-> Single-Thread games would show only a TINY improvement, because other running process' threads could be directed off the CPU the game is on, & sent to another of the 1-N cpu's present if they are less saturated & odds are? Because the GAME CPU is probably nearly so due to running a game?? They would be by the OS process scheduler subsystem kernel component)

However, as you can see above, for the reasons noted? Why not test SMP?? It shows SMP CPU-driven gains, not only video ones!

BUT, it makes THEIR CONCLUSION SOLID now (a 'partial truth' derived one), doesn't it? Of course it does... apk

gR3iF

10-13-2006, 01:50 PM

farcry is dual core? with the latest patch i have in mind correct me if necessary

and that dual core makes a huge difference
at least for me
i played cs:s with my amd athlon 3500+ clocked to 2,4gig , 1gig ocz ram 1:1@240@2-2-2-5 a x800xt pe and my raptors for a long time
now iam on my current system see specs

iam playing @1024x768 without all graphical effects so regarding to tomshardware the fastest card in this is a 9800xt

so graphic doesnt matter

BUT: the cpu matters iam feeling a slightly better performance when it comes to critical situations, like dogfight with heavy smoke and so on

its not that my 3500 was a dual core but its just that cpu matters in gaming and its a feeling if you have 35fps or 200fps

for me a huge difference

Alec§taar

10-13-2006, 02:11 PM

iam playing @1024x768 without all graphical effects so regarding to tomshardware the fastest card in this is a 9800xt

so graphic doesnt matter

BUT: the cpu matters iam feeling a slightly better performance when it comes to critical situations, like dogfight with heavy smoke and so on

its not that my 3500 was a dual core but its just that cpu matters in gaming and its a feeling if you have 35fps or 200fps

for me a huge difference

Agreed, & good point...

APK

10-13-2006, 02:15 PM

its not that my 3500 was a dual core but its just that cpu matters in gaming and its a feeling if you have 35fps or 200fps
for me a huge difference

can you see the difference between 35 and 200? :wtf:
you know that on cinema they use 24fps 'cause that is the limit for lagging

Alec§taar

10-13-2006, 02:21 PM

can you see the difference between 35 and 200?

I do when I am in the midst of a battle w/ particles of varied kinds flying, + smoke & explosions... (running Quake 4 SMP 1.3 "uncapped" via .cfg file hack) as Gr3if noted.

APK

P.S.=> Above all, this isn't directed @ Ashen Sugar specifically, he didn't do the test OR write that article: He just put it up here for us to look @ & voice our opinions + objections on it, as critique... & why!

And, since it is about SMP rigs + "INTEL vs. AMD" basically? I am really very happy having an AMD (even though MOSTLY I was INTEL in the past, & especially on SMP rigs (which a good majority % of my systems have been since 1996 or so))... & especially gaming: Fastest system I have ever owned is the current one, in my sig! apk

15th Warlock

10-13-2006, 05:25 PM

farcry not dual core/smp
serious sam 2 not dual core/smp
city of heros isnt benched its an mmo same with cov

cod2 isnt commonly benched by many sites as its not as popular as other games such as quake4/prey/doom3 and halflife2

Farcry, though not a dual core/smp enhanced game per se, takes advantage of dual core/HT systems when using Ati cards coupled with any Catalyst driver release after ver. 5.12:

Farcry receives the largest boost we experienced with our dual core testing a whopping 13% of an increase took the framerate from 81fps on average to 93fps on the Pentium D. The X2 was not far behind in terms of benefits with an average increase of 11% throughout our test levels and whilst not as impressive as the the 2 physical cores the P4 with HT still receives a nice increase of 5%.

http://www.driverheaven.net/articles/dualcore512/

I'm sorry, after downloading the latest Serious Sam 2 patch, I thought it had dual core support after reading the patch's release notes; I read them again today, and all the patch does is address some compatibility problems associated with HT/dual core systems. I'm sorry I mentioned it, it is my mistake :p

City of heroes can be benchmarked, here's a benchmark done by hardocp nonetheless:

http://www.hardocp.com/article.html?art=NjM3LDUsLDQw

The same happens with City of Villains, google for benchmarks and you are bound to found some:

http://www.anandtech.com/video/showdoc.aspx?i=2828&p=2

COD2 not too popular for benchmarking? :confused: I beg to differ, even hardocp uses it in benchmarks from time to time:

http://enthusiast.hardocp.com/article.html?art=MTAwMSw5LCxoZW50aHVzaWFzdA==

Some other COD2 benchmarks:

http://www.anandtech.com/showdoc.aspx?i=2668&p=8

http://www.firingsquad.com/hardware/call_of_duty_2_midrange_graphics/page5.asp

http://www.amdzone.com/modules.php?op=modload&name=Sections&file=index&req=viewarticle&artid=182&page=1

http://hardware.gotfrag.com/portal/story/33492/?spage=13

http://www.thechannelinsider.com/article/Intel+Core+2+Duo+Vs+AMD+X2+AM2Top+to+Bottom/188341_9.aspx

And finally, a comparison between X2 and FX AMD systems and Core 2 Duo systems using COD2 on a very respected hardware site:

http://www.tomshardware.com/2006/07/14/core2_duo_knocks_out_athlon_64/page12.html#games

In conclussion, a lot of SMP games were left out of the benchmarks in the article you quote, besides, what's the point of testing dual core systems if SMP games are not to be taken in consideration? Why not include those games?

Many current and future games take advantage of SMP coding, the upcoming Quake Wars, Crysis and Unreal 3 engine based games are glaring examples of game programers taking advantage of coding their games with SMP in mind.

gR3iF

10-13-2006, 05:29 PM

this is my fastest pc at all
and in my eyes i can see the difference i can even see if my monitor has 60 on refreshrate or 75

and i can view the difference between 30 and 50 fps and the difference between 50 and 100fps

u can see its just about you

how did i do this u can adjust the maximum fps in spurce via console
so for me its really clear i could upload you some videos and you can find out yoursefl if you want to

Gorre

10-13-2006, 05:30 PM

can you see the difference between 35 and 200? :wtf:
you know that on cinema they use 24fps 'cause that is the limit for lagging

Only in PAL.
NTSC uses 29.98fps

Alec§taar

10-13-2006, 05:34 PM

In conclussion, a lot of SMP games were left out of the benchmarks in the article you quote, besides, what's the point of testing dual core systems if SMP games are not to be taken in consideration? Why not include those games?

Great closing argument & point: Much like my own... I will tell you why, & it's only quoting their OWN words (again for the 3rd time):

Almost none of today’s games are performance limited by your CPU

Beg to differ - when Quake 4 went from single thread to multithread/SMP ready code? Well, see the photos above, & tell us it was NOT 'cpu constrained' being coded single threaded??

GPU <> everything in gaming & that proves it!

If you showed results like the ones I post above, using a real-world game no less, showing that statement is in BLATANT ERROR, on SMP utilizing games:

* Quake 4 SMP definitely shows CPU-DRIVEN gains... not just GPU ones.

:)

APK

15th Warlock

10-13-2006, 05:37 PM

magibeg

10-13-2006, 05:48 PM

Alec§taar

10-13-2006, 05:53 PM

24fps is the minimum needed for the human brain to perceive full motion, any framerate under that and you start noticing stuttering between frames, but your eyes can tell a diference between 24fps and 60fps, beyond that, I'm not so sure.

Extra 'frames-per-second' help during pitched battles where particles of blood & objects are flying, rocket trails & smoke animations etc. (special efx of weapons) matter as well..

APK

Gorre

10-13-2006, 06:05 PM

Extra 'frames-per-second' help during pitched battles where particles of blood & objects are flying, rocket trails & smoke animations etc. (special efx of weapons) matter as well..

APK

Fast games like Quake 3 FPS makes a massive distance.
Let's say that your screen is 24 cm wide (theoretically :P), and someone dives accross your field of view in one second.
That means you saw them move a centimeter at a time, over a second.
I suppose to a lot of people thats not important, but to the hardcore Q3 players I can see it making a massive difference.

15th Warlock

10-13-2006, 06:09 PM

Extra 'frames-per-second' help during pitched battles where particles of blood & objects are flying, rocket trails & smoke animations etc. (special efx of weapons) matter as well..

APK

Yes, those effects bring any card to its knees, the max average FPS you get on any benchmark, the more confident you can feel your system won't slow down during special effects intensive scenes. All these effects are very dependant on the CPU, just like A.I. scripts are. It's certain that in the near future we will see gaming programers taking advantage of two or more available cores, PPU cards and even a third GPU card for rendering these effects

That's why I like benchmarks that include both the min and the max FPS besides the average FPS. You don't want the min framerate to ever drop below the 25~30 FPS range.

Alec§taar

10-13-2006, 06:09 PM

Fast games like Quake 3 FPS makes a massive distance.
Let's say that your screen is 24 cm wide (theoretically :P), and someone dives accross your field of view in one second.
That means you saw them move a centimeter at a time, over a second.
I suppose to a lot of people thats not important, but to the hardcore Q3 players I can see it making a massive difference.

A gain = a gain, no matter what: I can use them!

:)

* CPU-driven gain is what we are debating here though... & I evidence that cleanly enough, & simply offered it as a counterpoint to the review noted by Ashen Sugar & 1 of their statements in particular, in this one:

Almost none of today’s games are performance limited by your CPU

(They really should have done a CPU-gain driven test like Quake 4 SMP 1.3 will show you does exist, but it is true, mostly it's your GPU/graphics card... but, they make it sound like DualCore does not help: Quake 4 SMP shows QUITE otherwise!)

ATI gains on SMP built Quake 4 beta (would be faster in final build, no debug symbols or code present in .exe final)

http://www.amdzone.com/pics/gaming/quake4/dualcore/x1800xl.jpg

NVIDIA gains on SMP built Quake 4 beta (would be faster in final build, no debug symbols or code present in .exe final)

http://www.amdzone.com/pics/gaming/quake4/dualcore/7800gt.jpg

* I can offer NO clearer proof... especially on SMP rigs (which this test compares) showing that multiple thread design DOES get you gains, even into the highest resolutions (where AA/AF are NOT as needed, especially AA)...

Those 2 screenshots of test results are the DIRECT proof of CPU-driven gain... period.

APK

10-13-2006, 07:31 PM

ocnoobert

10-13-2006, 11:20 PM

i upgraded my cpu and can now play games at full high specs that i could not play with my old cpu at all

10-13-2006, 11:25 PM

i upgraded my cpu and can now play games at full high specs that i could not play with my old cpu at all

wich was your old?

AshenSugar

10-14-2006, 12:01 AM

A gain = a gain, no matter what: I can use them!

:)

* CPU-driven gain is what we are debating here though... & I evidence that cleanly enough, & simply offered it as a counterpoint to the review noted by Ashen Sugar & 1 of their statements in particular, in this one:

(They really should have done a CPU-gain driven test like Quake 4 SMP 1.3 will show you does exist, but it is true, mostly it's your GPU/graphics card... but, they make it sound like DualCore does not help: Quake 4 SMP shows QUITE otherwise!)

ATI gains on SMP built Quake 4 beta (would be faster in final build, no debug symbols or code present in .exe final)

http://www.amdzone.com/pics/gaming/quake4/dualcore/x1800xl.jpg

NVIDIA gains on SMP built Quake 4 beta (would be faster in final build, no debug symbols or code present in .exe final)

http://www.amdzone.com/pics/gaming/quake4/dualcore/7800gt.jpg

* I can offer NO clearer proof... especially on SMP rigs (which this test compares) showing that multiple thread design DOES get you gains, even into the highest resolutions (where AA/AF are NOT as needed, especially AA)...

Those 2 screenshots of test results are the DIRECT proof of CPU-driven gain... period.

APK
res used was what? if its below like 1280 then i dont see it as valid since every gamer i know whos got a high end card plays at 1280 or 1600(or higher).

most of these kinds of cpu comparisons are run at lower res's to show cpu bound diffrances and have little to nothing to do with the video card, but the point is that most people who spend this kinda $ on a video card want to play at high res with AA and AF cranked up, benching otherwise is as pointless as skiping multi core games or patches for games.

AshenSugar

10-14-2006, 12:09 AM

I dont understand why they would bother writing that article. Sure the average gamer with their video card wont get a big gain from having a core2duo over an x2 but i feel the best way to discredit something like this would be to look at this in extremes.

System 1
amd fx-62 ($952.99)
voodoo2

System 2
pentium 2 300 (????$5????)
voodoo2

game test: need for speed 2

If you ran a comparison between those 2 systems now i'ld be willing to bet they'ld both be scoring around say 100fps. So with that logic according to the article i can safely say that a pentium 2 300 is 100% equal with an amd fx-64.
Its simply not a fair test! Now i threw in the price of the amd fx-62 from tigerdirect just for kicks but i hope you all get my basic idea of it.

almost any x2 made today can get fx62 perf with a little overclocking(on air mind you)

the point is that if the games gpu bound then it is valid to show people that theres no point in rushin out and grabbing core2 board/chip if you already have a a64 system, honestly most gamers with a64 systems would gain more from getting a x1900gt or 7900 card then by swaping out to a core2, thats what this was ment to show, only game i have seen thats cpu bound at all at 1600x1200 is oblivion, and its a pathetic pc port of a consol game, totaly unoptimized for pc, even at that the diffrance is you gotta turn shadows down 1-2 notches on an fx62 vs a x6800.

ofcorse people fail to take into account that you can get fx62 perf out of almost any current x2 cpu, even the $133usd x2 3600+ with a little SAFE overclocking.

never understood why intel people are so fixated on the fx line anyway, anybody got an answer for that?(seirously i dont get it)

AshenSugar

10-14-2006, 12:11 AM

24fps is the minimum needed for the human brain to perceive full motion, any framerate under that and you start noticing stuttering between frames, but your eyes can tell a diference between 24fps and 60fps, beyond that, I'm not so sure.

http://www.100fps.com/how_many_frames_can_humans_see.htm

read that, good artical about what the human eye can see and cant.

also its good to point out that motion blur in movies helps us not see the jumps from fraim to fraim at 24fps(normal movie speed)

if you pay attn you can see it in some scenes in most movies, but dont, once you start seeing it, well it drives you bonkers ;)

15th Warlock

10-14-2006, 05:28 AM

http://www.100fps.com/how_many_frames_can_humans_see.htm

read that, good artical about what the human eye can see and cant.

also its good to point out that motion blur in movies helps us not see the jumps from fraim to fraim at 24fps(normal movie speed)

if you pay attn you can see it in some scenes in most movies, but dont, once you start seeing it, well it drives you bonkers ;)

Thanks for the link, that was a very interesting read :)

Alec§taar

10-14-2006, 08:10 AM

res used was what? if its below like 1280 then i dont see it as valid since every gamer i know whos got a high end card plays at 1280 or 1600(or higher).

The tests above on ATI &/or NVidia cards is above, @ the resolutions you mention, showing CPU-driven gains (the exact same dualcore CPU on both, just diff. vidcards for comparison of SMP ready Quake 4) & again:

A gain's a gain, but this one's NOT according to what they're telling folks in the review you cited in your first post:

Almost none of today’s games are performance limited by your CPU

As you can see above - when Quake 4 went from single thread to multithread/SMP ready code? See the photos above, & tell us it was NOT 'cpu constrained' being coded single threaded??

GPU <> everything in gaming & that proves it!

If you showed results like the ones I post above, using a real-world game no less, showing that statement is in BLATANT ERROR, on SMP utilizing games:

* Quake 4 SMP definitely shows CPU-DRIVEN gains... not just GPU ones.

Their omission of testing only single-threaded games only tends to back that saying of theirs up, but the tests I show tend to make their statement less credible...

Bottom-line = That they should have tested SMP ready games, on SMP rigs.

most of these kinds of cpu comparisons are run at lower res's to show cpu bound diffrances and have little to nothing to do with the video card, but the point is that most people who spend this kinda $ on a video card want to play at high res with AA and AF cranked up, benching otherwise is as pointless as skiping multi core games or patches for games.

I said this before earlier: WHY ON EARTH USE AA/AF @ RESOLUTIONS ABOVE 800x600?

It's pointless to do, especially anti-aliasing: Imagery edges don't distort @ 1024x768 & above, you know... that's just overworking a videocard for NO good reason @ those res'...

APK

AshenSugar

10-14-2006, 08:37 PM

Alec§taar i use aa/af at 1600x1200 because some of the games i play have notable jaggies crawling on buildings/mountins/exct and i dont like it, i got my video card because i wanted to beable to crank up the settings for max quility and still get good fraim rates, and i can and do.

Most people i know who have good cards use aa/af at 1280 and higher reses because they see the same things i do.

try silkroadonline (google it) its a free forever mmorpg, once playing outside town watch the montins/hills without aa and af then exit the game and enable them, granted you have an nvidia card so you wont get the same quility i get but it will still show less gaggies/crawlies and better distance texture clarity.

AF is for any res, it clears texturs in the distance up, so why on earth anybody whos card can do it without perf loss wouldnt use it i couldnt understand.

in the past i would have agreed about AA at high res's but its been years since a good/decent card was choked by AA/AF at 1280 res and even my x800xt pe could do it at 1600x1200 in all but 2 games, quake4 and oblivion, and oblivion is so unoptimized its CRAZY, my game dev buddy bought it and is still waiting to see them properly patch it, from his prespective it should never have been releced in its current state(to many bugs to un-optimized)

oh and i loose 15fps to AA at most in games, so its worth it ^_^

http://www.ati.com/products/radeonx1k/imagequality.html

http://www.ati.com/products/radeonx1k/images/adaptive-aa.jpg
adaptiveAA (2d textures)is kickin as well, has nvidia got there version of it working properly yet?(seirous question, last time i saw it, well to be kind it was buggy)

http://www.ati.com/products/radeonx1k/images/af-quality.jpg
thats what AF can do for you, well me anyway, specly notable at HIGH res's

http://www.ati.com/products/radeonx1k/images/fsaa.jpg
fsaa, has the same effect even at high res's just less notable the higher you go, even my 9800pro was able to deal with it at 1280 res's in games of the day, and to good effect mind you :)

http://www.ati.com/products/radeonx1k/images/hdr1.jpg
and this is one you cant do, AA+HDR(fp16 hdr) as nvidia cards dont support aa+hdr at the same time, they use the same part of the card to do both ;)

read the first link, then check some reviews ^^

then read some other forums, most high end card owners use AA/AF at high res OR want to beable to at the very least.

personaly jaggies bugg the living chit out of me, crawling little jaggies...*grummbles* once you start noticing them they bug the shit out of you, like a scratch on your monotor b4 u started seeing/noticing them they didnt bother you, but once you do, well its all over, your gonna keep seeing them*grummbles*

af diffrances are easy to see when you pay attention, at least for me they have been.

http://www.ati.com/products/radeonx1k/images/fsaa.jpg

AshenSugar

10-14-2006, 08:43 PM

oh and i agree it would be nice to see a test of smp ready games, but most gamers got no clue what SMP/multi threding is, nor do they care, they play whats popular(sad but true :/)

and the jump in perf would almost certenly be leniar, as both chips are dual core, and the video cards the same, you likely wouldnt see anything but an across the board jump of a few fps on boath systems.

oh and please tell me that quake4 wasnt done on a dual core/HT nutburst system :P

Alec§taar

10-15-2006, 09:14 AM

AF is for any res, it clears texturs in the distance up, so why on earth anybody whos card can do it without perf loss wouldnt use it i couldnt understand.

This I noted above, AF I would POSSIBLY run (if needed that is), but AA I would not @ resolutions like 1600x1200... it's 'overkill' & taxes the vidcard needlessly.

oh and i loose 15fps to AA at most in games, so its worth it ^_^

There is where you & I differ: I don't GENERALLY take 'frames-per-second' losses as good, especially during 'pitched battles'...

fsaa, has the same effect even at high res's just less notable the higher you go, even my 9800pro was able to deal with it at 1280 res's in games of the day, and to good effect mind you :)

BUT, not @ 1600x1200, which is where AA is no longer needed imo @ least...

oh and i agree it would be nice to see a test of smp ready games, but most gamers got no clue what SMP/multi threding is, nor do they care, they play whats popular(sad but true :/)

Too bad they don't - because it does yield performance gains, CPU-Driven performance gains no less...

oh and please tell me that quake4 wasnt done on a dual core/HT nutburst system :P

You'd have to go to AMDZone (& other sites that tested) to see the exact machine specs used in their tests...

APK

AshenSugar

10-15-2006, 09:27 AM

15fps loss from 85+ isnt gonna be a problem even in pitched battles, and to me it makes things like sniping easyer, i am primarly a snarp shooter not an in your face brawler.

im good with hitscan weppons when needed but i prefer to stand back and use accurat range weppons the clearer things are at range the better i can get headshots ;)

the only thing i consistantly dissable is shadows, they IMHO are ugly and not realistic in current games anyway, just a perf drain, everything else i max out.

oh and from my exp with nvidia cards(pretty extencive) the perf losses to AA and AF are higher then with ati cards, duno why that is, but especly in d3d games theres a larger hit in AA/AF use, as to AF i dont see a fps hit in d3d or ogl using 9800,x800xt,x1900xt cards and infact a few times i have had the strange result of GAINING fps from cranking AF up(makes no sence but its true)

i would have liked to have seen farcry 1.4 beta patch with AA+HDR+AF enabled but they used a card that cant do HDR and AA at the same time(nvidia cards cant do it :( ) i have requested they do the same kind of test with more games using the latest avalable patches on 7900gtx and x1900/1950 xtx cards.

gR3iF

10-15-2006, 03:16 PM

sry man but that 15fps pitch may be noticeable
it hardly depends on the game you are playing i mean some games like bf2 are playable with some grahpics as long as you have 80+fps
but some other dont

15th Warlock

10-15-2006, 03:57 PM

adaptiveAA (2d textures)is kickin as well, has nvidia got there version of it working properly yet?(seirous question, last time i saw it, well to be kind it was buggy)

Yes it works just fine, in fact as of Detonator ver. 91.45 all 6 series cards support this feature along the 7 series.

Alec§taar

10-15-2006, 04:41 PM

sry man but that 15fps pitch may be noticeable
it hardly depends on the game you are playing

I tend to agree!

Plus, using AA @ resolutions like 1024x768, 1280x1024, & 1600x1200 (+ above them all if possible on your system) is pointless, since imagery borders @ those resolutions do not distort!

(Thus, using AA @ those higher resolutions while gaming is a waste of power making the vidcard overwork for no good reason, & loses you performance, needlessly).

This is one of the areas they are looking @ gaming on modern high-end vidcards that is erroneous imo, because you wouldn't use AA @ resolutions that high, & lose performance...

PLUS:

With the folks whose test Ashen Sugar put out? Via them NOT testing Quake 4 1.3 SMP (an SMP ready game mind you, no less)??

They tend to make this statement of theirs "true":

Almost none of today’s games are performance limited by your CPU

& these tests from AMDZone prove that is NOT true:

(They really should have done a CPU-gain driven test like Quake 4 SMP 1.3 will show you does exist, but it is true, mostly it's your GPU/graphics card... but, they make it sound like DualCore does not help: Quake 4 SMP shows QUITE otherwise!)

ATI gains on SMP built Quake 4 beta (would be faster in final build, no debug symbols or code present in .exe final)

http://www.amdzone.com/pics/gaming/quake4/dualcore/x1800xl.jpg

NVIDIA gains on SMP built Quake 4 beta (would be faster in final build, no debug symbols or code present in .exe final)

http://www.amdzone.com/pics/gaming/quake4/dualcore/7800gt.jpg

* I can offer NO clearer proof... especially on SMP rigs (which this test compares) showing that multiple thread design DOES get you gains, even into the highest resolutions (where AA/AF are NOT as needed, especially AA)...

Those 2 screenshots of test results are the DIRECT proof of CPU-driven gain... period.

APK

P.S.=> They did ONLY testing single-threaded games on SMP rigs for a test, & it's wrong!

(& even single-threaded games should show gains in their tests, but NOT as much CPU-wise - because the system CPU can direct their singlethread game engines to a less used CPU in an SMP setup, to gain SOME, & running other processes on the other CPU(s) present, giving the game running more of the 1-N cpu's present... but, as seen @ AMDZone, threaduse in game engines is shown as gaining frames-per-second MASSIVELY via multithreaded code design in a game engine in the AMDZone tests of Quake 4 SMP - PURE CPU-driven gains!)

(And, they said things about 1/2 truths-misinformation being used & lies on tests??)

APK

AshenSugar

10-16-2006, 12:45 AM

so they shouldnt test popular/common games only SMP games?

glad to hear they worked out the problems with Adaptive aa on the 6/7 seirse, the last time i tryed it on a 7 card it messed stuff up.

Plus, using AA @ resolutions like 1024x768, 1280x1024, & 1600x1200 (+ above them all if possible on your system) is pointless, since imagery borders @ those resolutions do not distort!
untrue, if you have a small monotor maby you dont see them, but with a 19in or larger you can see the jaggies on any current game.

few games to try
fable
silkroadonline
starsiege(old mech game)
NeverWinterNights

thats a short list of games that are EASY tests for jaggies, just LOOK for them as you move around, they are not hard to see my my 19in monotor, i could live with them as i did with my 9800at high res, but why when i can remove them fully and still keep VERY playable fraimrates?

oh forgot ut2k4 and unreal2 XMP(good tests for this kinda hing on most maps)

some people can live with jaggies, personaly i keep a high end card because i dont want to have to see them, im an image quility person, your clearly a FPS speed person, and thats fine, i tend to crank everthing to max, even editing the ini files to get quility settings the GUI leaves out due to hardware of the day not being good for it(u2xmp/ut2k4 have higher textuar quility settings then are listed, just gotta edit some lines and bam it looks better)

in ut2k4 at 1600x1200 4xaa 16xaf i get 90+fps constantly, thats more then playable to any reasonable person.

also in some games i use temporial AA at 2x giving the quility of 4x or better normal fsaa but with far lower perf hit, as long as i stay above 60fps it dosnt switch to non aa mode(not a problem in any game i have tested)

i understand you feel that stressing you card is wrong, i dont, i bought the card i bought so i could get max quility at high res with playable fraim rates.

oh and a 15fps DROP at 85+fps isnt gonna hurt my sniping or hitscan weppon use, honestly anything over 60 online and im happy, below that online and i change settings.

and yes we know there are gains from SMP but would the diffrance between amd and intel be that much diffrant if they both used the same code?, the likely answer is NO, it would be a liniar gain for both companys and the gfx card used would still be holding back the game from its max possable perf, the only way your gonna not be gpu limited in any test is to go x1950xtx crossfire or 7950gx sli(quadsli) or 7900gtx sli, but again, those are solutions that no normal gamer is gonna bother with, because most of us have figuared out that SLI and CF are not worth the bother, its better to just get a next gen card when they come out then to spend so much on 2 older cards in a vain attempt to hold off the need for a new card just a little longer.

do some polls on diffrent gaming forums, see how many people who buy 79xx and x19xx cards want to beable to use AA/AF at 1024/1280/1600 reses, you will be supprised that most of them at least want to beable to, i mean u spend this kinda bank on a video card you want the works, in my case its aa+hdr+af in oblivion at 1280 res(1280 because oblivions poorly optimized(being kind here) and runs like crap on the pc.

ps no i dont just play oblivion but it was an example, i want to beable to play games with all fetues ENABLED at higher res's with good/resonable/fast fps :)

AshenSugar

10-16-2006, 12:53 AM

my buddy wanted me to ask if your saying that games arent gpu limmited at all just cpu limited due to lack of SMP code?
he thinks thats what your saying anyway, and hes a game dev, he would like you to clarify if you think that quake4 would give better FPS in smp mode with a better video card or not.

sorry but hes not got an account here and hes stainding over my sholder wanting me to post(told him he could make an account and post but he wouldnt lol)

hes just sure your saying that the only reasion games dont show HUGE gains with core2 and current video cards is because they arent SMP enabled.......

anti-liberal

10-16-2006, 04:40 AM

just :nutkick:the kids and teach them they are spending too much money, crossfire is in my opinion useless, a single x1900xtx can play all games at max settings above 35fps :shadedshu

I'm sorry but I got to LAN tournaments and what not and a Professional Gamer and i'm not looking for the best graphics... I'm looking to get 333 FPS in Call of Duty 2 or 100 in BF2. I have put over a grand into my system and I can't get to 333fps in cod2 with medium graphics

AMD 3800+ OC 2.7ghz
1.5 GB DDRII G.SKill 667 MHZ Timings 4-4-4-12 (about to upgrade to 2gb)
WD 80 GB Sata 7200RPMS
x1800xt 256mb 256bit (I am about to sell to get a 7900gto... I'm not looking to spend too much cause dx10 cards will be out soon)
MOBO ASROCK
Antec 550 Watt PSU SLI ready

10-16-2006, 10:31 AM

Alec§taar

10-16-2006, 12:24 PM

so they shouldnt test popular/common games only SMP games?

They should have tested BOTH, not just single-threaded games... their testing ONLY THOSE (single-threaded games) lends TOO MUCH 'credence' to what they said above in this statement:

Almost none of today’s games are performance limited by your CPU

The tests @ AMDZone whose photos I used showed quite otherwise, when Quake 4 was tested in single cpu ready builds, vs SMP builds (beta thru 1.2-1.3 builds)... where a CPU-Driven gain was CLEARLY evidenced, period, & the fact Quake 4 was literally "CPU Constrained" by LACK OF SMP READY CODE!

untrue, if you have a small monotor maby you dont see them, but with a 19in or larger you can see the jaggies on any current game.

I have a 19" Trinitron, & once I get to 1024x768, jaggies don't occur as much or noticeably as they do @ resolutions like 640-480-800x600... get up past that, into the 1280x1024-1600x1200 (or more)? They're gone...

No need to use Anti-Aliasing @ those resolutions up to & past 1024x768.

your clearly a FPS speed person, and thats fine, i tend to crank everthing to max, even editing the ini files to get quility settings the GUI leaves out due to hardware of the day not being good for it(u2xmp/ut2k4 have higher textuar quility settings then are listed, just gotta edit some lines and bam it looks better)

I do the same, for a BALANCE of speed & looks... game config file editing.

i understand you feel that stressing you card is wrong

Needlessly, by running AA @ resolutions up to & past 1024x768... yes, I do.

oh and a 15fps DROP at 85+fps isnt gonna hurt my sniping or hitscan weppon use, honestly anything over 60 online and im happy, below that online and i change settings.

It will me (and others such as Gr3if who noted this also) in a 'pitched battle' & slow me down enough to probably get 'scorched'...

and yes we know there are gains from SMP but would the diffrance between amd and intel be that much diffrant if they both used the same code?, the likely answer is NO

I'd take a look above, & see what diff.'s the Quake 4 SMP code made - it actually got gains LARGER on ATI cards, than it did on NVIDIA cards... it MIGHT be possible you'd get better scores using either INTEL or AMD CPU's (heck, might be better/worse w/ diff. chipsets & mobos used)...

do some polls on diffrent gaming forums, see how many people who buy 79xx and x19xx cards want to beable to use AA/AF at 1024/1280/1600 reses, you will be supprised that most of them at least want to beable to

It's pointless to do so though...

Especially using AA @ res' up to & past 1024x768... there is little to NO NEED for it the higher up in resolution you go, again, because jagged edges on imagery begins to disappear @ those higher res ranges.

APK

Alec§taar

10-16-2006, 12:29 PM

Take a read there Ashen Sugar:

I'm sorry but I got to LAN tournaments and what not and a Professional Gamer and i'm not looking for the best graphics... I'm looking to get 333 FPS in Call of Duty 2 or 100 in BF2.

Seems to me that a "real gamer" is talking there, & what is he looking for? SPEED... just as I do!

:)

* BUT, to each his own, on that account, when you come down to how you use your system & what you want out of it... but, his statement DOES make my point stronger, since you said I ought to ask some "real gamers" what-is-what on this account:

(I.E.-> Using GFX cards more for speed, or quality of view settings in their drivers settings, or gaming config files tweaks...)

my buddy wanted me to ask if your saying that games arent gpu limmited at all just cpu limited due to lack of SMP code?

No, elements of both exist... but, the site you took that from, by omitting testing SMP ready game code, & testing single-threaded code only?

Well, in doing so?? They told some 'falsehoods', & they BITCHED ABOUT THAT IN THE ARTICLE NO LESS... (lies in articles, etc.) with this statement:

Almost none of today’s games are performance limited by your CPU

Quake 4 SMP beta's (would be faster in final builds, again, due to debug symbols & err trap code being gone) show QUITE OTHERWISE above... pure CPU-driven gains resulted using the SMP ready builds (vs. single thread build) of Quake 4, period.

APK

P.S.=> Tip for you, anti-liberal? I wouldn't use Anti-Aliasing @ FULL OUT/FULL BORE settings @ resolutions up to & past 1024x768... it's over-working the graphics card doing the AA processing into that high of a res. range... needlessly! Jaggies nearly disappear by 1024x768, & most certainly do past that... apk

AshenSugar

10-16-2006, 01:11 PM

I'm sorry but I got to LAN tournaments and what not and a Professional Gamer and i'm not looking for the best graphics... I'm looking to get 333 FPS in Call of Duty 2 or 100 in BF2. I have put over a grand into my system and I can't get to 333fps in cod2 with medium graphics

AMD 3800+ OC 2.7ghz
1.5 GB DDRII G.SKill 667 MHZ Timings 4-4-4-12 (about to upgrade to 2gb)
WD 80 GB Sata 7200RPMS
x1800xt 256mb 256bit (I am about to sell to get a 7900gto... I'm not looking to spend too much cause dx10 cards will be out soon)
MOBO ASROCK
Antec 550 Watt PSU SLI ready

u spend over 1k on that, HAHAHAHA, thats funny, asrock mobo, antech psu, 1.5gb ddr2, 80gb hdd, x1800xt 256mb, my OLD system was at least as powerfull as that and it was socket 754!!!!

first off i would like to note that dx10 cards are only gonna come out after vista is avalable, and then you will be FORCED to use VISTA to run dx10 games as ms dosnt currently plan dx10 support for anything but vista, vista needs 2gb ram MINIMUM to perform at "reasonable" levels, it needs 3-4gb to be decent for gaming, most game devs are not gonna rush to use dx10 because most gamers arent gonna be running vista for a couple years after it comes out.
if you dought this check games like halflife2 where they have dx7 and 8 codepaths dispite the fact that dx9 cards had been on the market for years when it came out, game devs run behind the "top thing" because the majority of gamers run older hardware such as dx9 cards.

and if your looking for dx10 complyance, your taking a step BACK by moving to a 7900gt, the x1k cards are closer to dx10 then the 7 seirse;) and you can get a x1900gt based on the x1950core cheaper overclock it and BAM faster then 7900 in most tests, oh yeah and you can use AA+hdr with them, oh wait, your about the FPS so you wouldnt care about hdr or aa, you would be turning everything to forced off and down anyway....ROFL.

Alec§taar
i dont DROP below reasonable FPS even is "pitched battles" thats the point, my minimum fps is never beloe 70/75 , thats plenty for me to stay alive/not get scorched.

i know nvidia has lower minimum fps then ati in the x1k range of cards so your probbly correct not to use AA/AF/any advanced fetures with your card at decent res's because it would cause you to drop below a reasonable fraim rate when things get intence.

my system is in my "my system" link if your wondering what i currently run, i would say im more a PowerGamer then the so called "pro gamer" above is, he to me comes across as a CS kiddie, "its all about getting 333fps"

333fps is just stupid, in reality your better off locking to refresh (vsycn enabled) because taring DESTROYS accuracy PERIOD.

im not a "pro gamer" i play for FUN not PROFIT, pro=see it like a job and to me that takes the fun out of it.
i did once win 150bucks on a gaming turny, it was fun, but i wouldnt want to do it alot, because the presure to win removes the fun, going "pro" takes the fun out of it for me.

building a system for gaming is about balance, just like overclocking is about balance, you gotta know what to get and how to configuar it for best performance and quility/stability.

i personaly am very happy with my online games never droping below 70fps, my refresh rate is 85hz and i use vsync enabled other then for testing because honestly higher fps then the screen can show are POINTLESS and acctuly make it harder to game(taring is nasty buisness)

well time for some sleep its 5am :P