3DAnalyser - Running Advanced Games on GF4 MX...

[Synth]

I knew about 3dAnalyser a long time now but avoided suggesting it here due to its hardcore nature.

That said, I happened to come across this thread on getting POP to work on GF4MX. These identical settings should work on most advanced games with GF4MX...

http://www.techjamaica.com/forums/showthread.php?t=4866                    

[LiQuiDSNaKe]

I have a Gf4000 not the 440 which i think would have been the better one to buy.                    

[Synth]

gf4000 is hardware identical to 400(real) or 440.

Its just basically higher clocked + agp8x support == better.                    

[LiQuiDSNaKe]

thanks but i'm gonna check to see if i can get a radeon first or something.Thanks alot though.                    

[Crixx_Creww]

before reading the article( cause iam not in the mood)

can i use these said tweeks or watever to maybe juice my 5200?
                     

[TrinireturnofGamez]

before reading the article( cause iam not in the mood)

can i use these said tweeks or watever to maybe juice my 5200?

  no , what this does is make the GF4 emulate pixel shaders  , and tell POP that i can be run . Your 5200 already has pixel shaders , so doing this would just make it slow as hell

  nvidia nvidia ....  if yall had gotten an 8500 or a 9000 in d first place, none of this would have happened, besides, the 8500 is faster than a 5200 ultra , 9600 non pro, and 5600/5700 XT/non pro .  And it has been available since 2001                      

[Synth]

Saying its faster is WRONG unless you specify in what it is. It doesn't support DX9 so in DX9 games which is backward compat to dx8.1, its unfair to say an 8500 is faster just because by default in dx8.1 mode it runs faster than a 9600 np in dx9 mode...By mode I mean same shader version, etc.
When making comparisons for "best" care must be taken to compare apples to apples and not oranges.

That said I severly doubt a 5700XT can be beaten by a 8500 in ANY mode.

BTW 8500 equiv is 9100 not 9000. 9000 is hardware raped.

8500 was ATI biggest technical turning point. It is still competitive on modern games today, once you don't mind running in DX 8.1 mode.                    

[TrinireturnofGamez]

9100 is also raped hardware , actually the 9000 pro is the fastest DX8 card ATI made after the 8500 , the 9100 pro , wich was the 8500 with even HIGHER clocks and higher bandwidth, was cancelled, it would have undercut sales of the 9600s  .  
   
 DX 8.1 mode gives just as good quality as DX9 mode in most games , with the exception of Farcry and a handful of others .  the DX9 support on all FX cards is poor , it supports all the shaders and high accuracy, but it cannot render high levels of shading properly , so  in truly extensively shaded games [eg. HL2 ], the FX cards have to resort to DX8 levels of shading anyhow .  Do not jump and say i am talking crap, RESEARCH .

 I am pretty sure an 8500/9600 np will pwn a 5700xt , XT= crap for nvidia, the XT cards are all slower than the non ultra models ,
                           

[Synth]

9100 is also raped hardware , actually the 9000 pro is the fastest DX8 card ATI made after the 8500 , the 9100 pro , wich was the 8500 with even HIGHER clocks and higher bandwidth, was cancelled, it would have undercut sales of the 9600s .

ATI Radeon 8500 275 4 1100 2 8 2200 550 128 8.6
ATI Radeon 9000 Pro 275 4 1100 1 4 1100 550 128 8.8
ATI Radeon 9100 275 4 1100 2 8 2200 550 128 8.8
ATI Radeon 9200 250 4 1000 1 4 1000 400 128 6.4
ATI Radeon 9200 Pro 300 4 1200 1 4 1200 600 128 9.6
ATI Radeon 9600 325 4 1300 1 4 1300 400 128 6.4
ATI Radeon 9600 Pro 400 4 1600 1 4 1600 600 128 9.6
NVIDIA GeForce FX 5200 Ultra 325 4 1300 1 4 1300 650 128 10.4
NVIDIA GeForce FX 5700 Ultra 475 4 1900 1 4 1900 900 128 14.4

How is a 9100 "raped"? The fact is ATI just rebadged the 8500 and gave it a 9100 rating to fit into the current naming scheme. It reflects the fact it is indeed superior to the hardware raped 9000 series. They both use identical RV200 cores. Even though the 9000 series uses a more efficient core (RV250) they did "rape" it and it can only do 1 texture per pipeline compared to 2 in 8500/9100. The clock speed boost over the 9100 cannot compensate enough in most games (especially if newer) making the 9100 faster in DX8.1 games than even the 9200pro. In any case the 9200pro is a DX9 classed product so you can't directly compare in this case.

DX 8.1 mode gives just as good quality as DX9 mode in most games , with the exception of Farcry and a handful of others . the DX9 support on all FX cards is poor , it supports all the shaders and high accuracy, but it cannot render high levels of shading properly , so in truly extensively shaded games [eg. HL2 ], the FX cards have to resort to DX8 levels of shading anyhow . Do not jump and say i am talking crap, RESEARCH .

If DX8.1 is just as good they wouldn't have bother to make DX9 mode would they? Its not up to YOU to decide whether its "good enough". Please try and focus now.

The argument is based on this and this alone:
"the 8500 is faster than a 5200 ultra , 9600 non pro, and 5600/5700 XT/non pro"

I am saying your logic is FLAWED since you are comparing apples to oranges. Ie dx8.1 mode to dx9 mode. Now had you said DX8.1 mode for BOTH and had benches to prove your point then we would not be having this "argument". How did FX get into all this? Your bringing in more and more variables which have nothing to do with your original point.

Just a reminder...

Your trying to prove than an 8500 is "faster". Course up to now you haven't actually specified "in what" its "faster" in.


Research you say? All the research in the world won't help if one does not know what they actually researching and what for.                    

[TrinireturnofGamez]

In any case the 9200pro is a DX9 classed product so you can't directly compare in this case.

 

 the 9200 is still DX8.1 class hardware  ,   wherever you heard otherwise is an unreliable source of info .  Most of the ' GPU cheat sheets' are erranous .

    The 9100 is raped because it can't make 8000 3dmarks without serious O/Cing on a 2.8 ghz P4 ,  those statistics mean absolutley nothing in real world , the 8500 has higher pixel filling power than the 9600 pro , but who wins overall? the 9600 .  The X700 has alot less memory bandwidth than the 9800XT yet it manages to own it all the time .  
  Though i have never seen benchmarks of the 5700 XT ,  find me some so i can compare it to my own performance .  
 
  the 6800 is a DX 9.0 C class product, yet it is DIRECTLY compared with the  X-800s which are only 9.0 B .  Even though the 8500 may be an apple , and the 5200 an orange , i believe the apple to have been the better buy , last year . I myself had to choose between an 8500 , and a 5200 for my upgrade last year, and i am very happy with my choice .                      

[Synth]

the 9200 is still DX8.1 class hardware , wherever you heard otherwise is an unreliable source of info . Most of the ' GPU cheat sheets' are erranous .

It does not have full DX9 support. But it has some.
Whether practically or not its "DX9 classed" is another matter altogether. A matter which has nothing to do with original question.

The 9100 is raped because it can't make 8000 3dmarks without serious O/Cing on a 2.8 ghz P4.

Raped in hardware world mean hardware is removed/limited from original core/chipset. 9100 is not "raped" in any way. Its just that it typically it comes

slower clocked.THATS THE ONLY DIFFERENCE. Slower clocked CANNOT mean RAPED. Not all cards in same generation are ever clocked the same. But it would

non-sensical to call them all RAPED. Raped has nothing to do with benchmarks either since its entirely possible from one manufacturer to the next a card

might be so highly clocked it gives higher benchmarks.

Those statistics mean absolutley nothing in real world , the 8500 has higher pixel filling power than the 9600 pro , but who wins overall?

Those stats were just to prove my point about hardware difference. Did I ever say "higher pixel filling power" is the ONLY THING that equates to overall

performance? No, so stop assuming.

The X700 has alot less memory bandwidth than the 9800XT yet it manages to own it all the time .

The X700 DOES NOT OWN The 9800XT ALL THE TIME. Get wider sources. The main reason an X700 has a chance at all is because its clocked so much higher.

Though i have never seen benchmarks of the 5700 XT , find me some so i can compare it to my own performance .

Anyone else finds this amusing? Using a mystical card which he has seen no benches of as "evidence". The reason I asked YOU was because such a card DOES NOT

EXIST...at least not ever for mass market.

the 6800 is a DX 9.0 C class product, yet it is DIRECTLY compared with the X-800s which are only 9.0 B . E

Dude you saying 9.0c vs 9.0b is anywhere close as comparing 8.1 to 9.x? LOL. They can compare those mostly because they are from same generation. The .c

additions are MINOR.

even though the 8500 may be an apple , and the 5200 an orange , i believe the apple to have been the better buy , last year . I myself had to choose between

an 8500 , and a 5200 for my upgrade last year, and i am very happy with my choice .

I never said it wasn't a good buy in fact I do own a 9100 which I love.

Geez...

You constantly ignore most of my direct points/questions. And you keep adding new "facts" which has nothing to do with original question.
And you claim my sources are inaccurate without proof.

This thread is headed for "closed" status.                    

[TriniXaeno]

lol

passion causes such things.

be mindful of that.

Passion is Power.

Round 2! Fight!                    

[vinion2000]

in the ati fanboy defence i think that in when he said 5700xt i think he meant the 5700le which does suck ass. the gf4ti 4200 surpasses it since like you said its a dx9 card.

plus doesnt the 9200 also has issues with PoP. ive seen this twice personally.

also the 8700 suck ass compared to the 5700. you need to read some more. cause even gf4 can do better at playing HL2 than most cards but thats case it doesnt process all dx commands and the game looks horrible.                      

[TrinireturnofGamez]

no such thing as an 8700 ! , if you mean 9700 , you on serius crack cause that can beat up on a 5900 ultra sometimes . If you mean the Fire GL 8700 , its not MEANT to play games , its meant to render 3d for CAD , movies etc.  Whoever has that in their GAMING only rig, needs to check a physcologist .

  All FX cards are forced to play HL2 in DX8 mode anyhow, just like the Geforce 4 TI [ which is still faster than them in DX8 mode ] .

    Can someone with a 5700 xt tell me how much 3dmarks they score? cause i have seriously NEVER seen a benchmark or a review of it, just an occasional mention of its existance .  

 I'm not saying to buy an 8500 over a 5700 , I'm saying that if you people had gone @ti last year/ year before , you would have payed 100 USD for a card that will perform VERY well even now .

 
                         

[coldstorm]

i Sure synth said in he last post the 5700 xt does not exist to mass public. Very well is opiniated answer since the 8500 doesn't have dx 9.0 support and some people like to game with full detail rather than missing detail.                    

[cereal_killer]

would the 3d anyliser turn on shadows in halo with my mx 4000?                    

[Synth]

Even if it could it would just slow the game down even more since it be software and not hardware rendered. An no amount of cpu power will be able to compensate.

This is simply meant to trick the game into thinking features are there so it can work. How well it works vary from game to game and features it has to "emulate".

The more it has to trick/emulate the more cpu power will be used...and the slower the actual game gets in terms of fps OR the crappier the game will look.
                     

[TrinireturnofGamez]

the Geforce MX4000 is a dx7 card , and has to use the  'fixed function' codepath in halo . No effects  are possible [other than sprites]  . Thats another sucky thing about the halo engine , lack of adaptability .
   DX8 is alot better than most people might think , you can still get all the pretty water effects , soft shadows etc. , all Geforce FX cards default to DX8 mode in half life 2   cause Vavle is Nvidia hating , and FX cards have are weak at shading .  
   The difference between DX8-DX9 is almost the same as DX9-DX9c , just more shader instructions per pass , and a few extra features that no one uses .