In House HD2900XT vs. 8800GTS 640

Page 24 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BFG10K

Lifer
Aug 14, 2000
22,709
2,995
126
I don't mind if the vehicle sequences are third person, it's mainly aiming the regular guns that I'm conerned about.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: BFG10K
I don't mind if the vehicle sequences are third person, it's mainly aiming the regular guns that I'm conerned about.

OK, in the CP there is a check-box that enables "FPS View"
-i imagine you can just keep it that way ... i doubt it unchecks itself

i haven't played LP for more than an hour or so ...
--ask me anything about Overlord
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
The demo for Lost Planet is a good enough reason to steer clear of this game... Tried on three different setups and each one has it's own unique odd issue. One of them was the menu's acting in turbo mode, so I was unable to make changes to the settings except based on luck. The second one locks up right as the big bug comes out of the second demo and the third one? Can't remember the issue with the third one. What a crappy port. If you can get the game to work, awesome. But the game is a buggy port and the performance on today's graphics cards just plain blow.

Do yourself a favor BFG and download the demo. If you can play through BOTH maps until the end, it would be safe to buy. Otherwise, steer clear of this POS port.

Edit - And not to hi-jack the thread, but what is the hell is going on with these ports? All the tom clancy games are crap ports, this new capcom crap port, TDU was a relatively crappy port... I suppose as long as people buy them (looks at Apop) they will continue to make them under-par
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
the full game plays MUCH better then the Demo, AA
- very playable on my system at 11x8, everything hi/med - 0AA/8xAF ... and the nvidia cards play it much better ... i am waiting for Cat 7.7 before proceeding.

after BFG10K asked about FP view, i went ahead and played it again for awhile - switching back and forth

it looks great either way ... FP view is easier to shoot and aim ... thrid person has a lag with the camera as you swing your [especially big] gun around ... so, i guess i would also prefer FPS

--and too bad for you AA ... LP is NOT a crappy port ... it is a HELL of a lot better than the xbox360 version in every way and a decent game in its own right and with xbox360 you are *stuck* with 3rd person

Capcom really surprised me with this one ... after the Crap that was the RE-4 port ... KB+Mouse rules in LP
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
If I have time later, I can link several reviews that consider this a less than desireable port, which lines up with my experience.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: ArchAngel777
If I have time later, I can link several reviews that consider this a less than desireable port, which lines up with my experience.

screw the reviews ... i read them also

RE-4 IS a Crap port ... it puts the Crap in Crapcom ...
otoh, LP:EC is a very good game and it is SO obviously not a "straight port" ... they gave some real thought to the KB+M instead of forcing me to use my xbox360 controller
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
The game on the 360 is awesome... You might consider the port acceptable, but I don't at this point. I think we can BOTH agree that the port could be better, yes? Or do you consider the port near perfect?
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
yes, better than the 360 version - "more awesome"
... a decent port ...
-unlike a bad port

every game i have played "could have been better", imo. :p

-just my opinion ... which is different from your opinion in this case
.. the full game plays much better than the demo - and it is patched
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Doesn't Lost Planet take advantage of Quad Core CPUs? It might be cool if you could disable one of the cores from the CPU to see how much difference having the extra core makes.

Are you guys gonna do a more elaborate writeup and a conclusion of some sort?

Also, I'd like to see C&C 3 and Supreme Commander benchmarks. :)

It looks to me as though the two cards came out pretty equal in your tests. I'm thinking some overclocking would be great, and could in fact be the main difference between the two cards.

Out of curiousity, do you know if 64-bit Vista would have an impact on performance?
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
SB,

I tested with the demo between dual core and single core, the difference was about 10%. I made sure to test in a CPU limited scenario. This is with my X2.

Apop,

So you said the final game is better? I am gonna trust you bro... Seriously... So, now I gotta ask you. Does the game PERFORM better than the demo? Or is the demo a realistic benchmark as far as the final version of the game is concerned?
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: SickBeast
Doesn't Lost Planet take advantage of Quad Core CPUs? It might be cool if you could disable one of the cores from the CPU to see how much difference having the extra core makes.

Are you guys gonna do a more elaborate writeup and a conclusion of some sort?

Also, I'd like to see C&C 3 and Supreme Commander benchmarks. :)

It looks to me as though the two cards came out pretty equal in your tests. I'm thinking some overclocking would be great, and could in fact be the main difference between the two cards.

Out of curiousity, do you know if 64-bit Vista would have an impact on performance?

i believe LP is Dual-Core optimized

as to more testing ... yes, i AM planning to redo all my HD2900xt benches for Cat 7.7 ... of course, i won't have a 8800GTS to compare with .. but maybe Keys will update his GTS scores when nvidia releases the next certified ForceWare
--and i plan to also repeat everything with CrossFire ... if i get a 2nd 2900xt

... Keys and i are discussing "in-house" testing as a 'regular' feature

C&C 3 and Supreme Commander benchmarks.
they don't interest me to buy the games ... if there is a free demo benchmark, i will be glad to test it for you
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Originally posted by: apoppin
Originally posted by: SickBeast
Doesn't Lost Planet take advantage of Quad Core CPUs? It might be cool if you could disable one of the cores from the CPU to see how much difference having the extra core makes.

Are you guys gonna do a more elaborate writeup and a conclusion of some sort?

Also, I'd like to see C&C 3 and Supreme Commander benchmarks. :)

It looks to me as though the two cards came out pretty equal in your tests. I'm thinking some overclocking would be great, and could in fact be the main difference between the two cards.

Out of curiousity, do you know if 64-bit Vista would have an impact on performance?

i believe LP is Dual-Core optimized

as to more testing ... yes, i AM planning to redo all my HD2900xt benches for Cat 7.7 ... of course, i won't have a 8800GTS to compare with .. but maybe Keys will update his GTS scores when nvidia releases the next certified ForceWare
--and i plan to also repeat everything with CrossFire ... if i get a 2nd 2900xt

... Keys and i are discussing "in-house" testing as a 'regular' feature

C&C 3 and Supreme Commander benchmarks.
they don't interest me to buy the games ... if there is a free demo benchmark, i will be glad to test it for you
If you guys want some help benchmarking, I don't mind. I have an 8800GTS 320mb and within a week or so I'll have a Q6600 with 4GB of RAM.

C&C 3 is the best game I have played since HL2. It has a demo, and so does SC.

I look forward to the Cat 7.7 benches.

Apparently the 2900 cards are faster than anything in Crossfire. Good luck. :)
 

superbooga

Senior member
Jun 16, 2001
333
0
0
LP isn't dual-core optimized. It's multi-core optimized. There's a setting in the options that lets you adjust the number of simultaneous tasks supported.

There's a very interesting article in Japanese somewhere that talks about how Capcom designed the game engine. It's wasn't designed for the X360. Instead it was designed for multi-processor systems in general.
 

videopho

Diamond Member
Apr 8, 2005
4,185
29
91
Frankly I'm getting a bit frustrated with my new 8800gts's crappy performances, especially with my fav game: FSx.
I'd love to see a benchmark done on this game in order to see how the 2900 stack up against the 8800's.
I currently use the latest signed driver from Nvidia at the moment (whatever it is 16x.x).
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
SB,

Just noticed today that Anand tested that function just today (I think). Check it out here The last graph shows the CPU limited map (The Cave) and let me tell you, the game is HEAVILY optimised for QC by the looks of it.

There are now two games that really benefit from QC... More to come, I am sure...
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: SickBeast
Originally posted by: apoppin
Originally posted by: SickBeast
Doesn't Lost Planet take advantage of Quad Core CPUs? It might be cool if you could disable one of the cores from the CPU to see how much difference having the extra core makes.

Are you guys gonna do a more elaborate writeup and a conclusion of some sort?

Also, I'd like to see C&C 3 and Supreme Commander benchmarks. :)

It looks to me as though the two cards came out pretty equal in your tests. I'm thinking some overclocking would be great, and could in fact be the main difference between the two cards.

Out of curiousity, do you know if 64-bit Vista would have an impact on performance?

i believe LP is Dual-Core optimized

as to more testing ... yes, i AM planning to redo all my HD2900xt benches for Cat 7.7 ... of course, i won't have a 8800GTS to compare with .. but maybe Keys will update his GTS scores when nvidia releases the next certified ForceWare
--and i plan to also repeat everything with CrossFire ... if i get a 2nd 2900xt

... Keys and i are discussing "in-house" testing as a 'regular' feature

C&C 3 and Supreme Commander benchmarks.
they don't interest me to buy the games ... if there is a free demo benchmark, i will be glad to test it for you
If you guys want some help benchmarking, I don't mind. I have an 8800GTS 320mb and within a week or so I'll have a Q6600 with 4GB of RAM.

C&C 3 is the best game I have played since HL2. It has a demo, and so does SC.

I look forward to the Cat 7.7 benches.

Apparently the 2900 cards are faster than anything in Crossfire. Good luck. :)

of course .. the more the merrier ... and thanks ... i also need a booster PS [i think ... 850 w is cutting it pretty close for 2900xt X-Fire]

are these demos set-up with a "built-in" benchmark like CoJ and LP?
--often as in the case of BOTH CoJ and LP, the game runs better then the demo [having been patched]
=================
Originally posted by: superbooga
LP isn't dual-core optimized. It's multi-core optimized. There's a setting in the options that lets you adjust the number of simultaneous tasks supported.

There's a very interesting article in Japanese somewhere that talks about how Capcom designed the game engine. It's wasn't designed for the X360. Instead it was designed for multi-processor systems in general.
i really haven't played it enough to recommend it ... what i played, i like

i don't like the "checkpoint only" saves ... but it is kinda easy ... so i don't know ... i don't HATE console ports like some of you [AA] - except for obvious straight ports like RE4; clearly Capcom didn't take full advantage of the PC platform - but it DOES have a FPS view and good support for KB+M [although, to be fair - it looks like they left an "auto aim" in].

Perhaps i am just "starving" for DX10 games that make it more enjoyable ... and Overlord IS more fun ... i am playing that one now [great looking DX9c]

the reviews all mentioned that it is repetitious ... that you are 'stuck in the snow' for almost all of the game
Perhaps i like it because i am stuck in the desert
:sun:

 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Originally posted by: apoppin
of course .. the more the merrier ... and thanks ... i also need a booster PS [i think ... 850 w is cutting it pretty close for 2900xt X-Fire]

are these demos set-up with a "built-in" benchmark like CoJ and LP?
--often as in the case of BOTH CoJ and LP, the game runs better then the demo [having been patched]
I don't think they have built in benchmarks, so I'm not quite sure how we would have to do it. I think if we play back a replay it might do the trick; I'm thinking the more bots the better to see how multi-cores scale in both games (although I don't think it has an effect in C&C 3).

What's a 'booster PS'? I have pretty much the same one as you and it's been fine for me, although I haven't really loaded it too much yet. I'm a little discouraged to hear that you've managed to max it out already...
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
http://www.totalvideogames.com...dsec=download&did=7166

this SC demo? ... it is 1.1GB :p
--is there a built-in bench?

if NOT, then we need to find or create a custom time-demo that we both could share

-BtW - this is a 10 overnight D/L on 56K for me ... and my D/L manager better get it right
What's a 'booster PS'? I have pretty much the same one as you and it's been fine for me, although I haven't really loaded it too much yet. I'm a little discouraged to hear that you've managed to max it out already...
i haven't maxed it - it is fine for oc'ing a single HD2900xt; maybe - just maybe for stock 2900xt Xfire ... but i am missing the extra cables and connectors ...
--a booster PS provides dedicated power for the 2nd GPU and over-clocking options

EDIT:
http://www.bjorn3d.com/read.php?cID=1099&pageID=3415

according to this, SC full game has a built-in benchmark
--and also according to them, the GPU is less important then the CPU
Supreme Commander

- 4xAA/16xAF
- Everything set to the highest quality setting
- Built in benchmark used. The benchmark runs a map with 4 opponents that blast eachother. It runs for about 7.5 minutes.

The HD2900XT performs pretty similar to the GeForce 8800GTS, beating it with a few frames at the higher resolutions. Looking at a graph of the actual fps during the whole benchmark, recorded with fraps, it is obvious that there really is not much difference between the different cards in this benchmark. The game relies heavily on the CPU and it is just in a few instances that you can see a definitive difference between the cards.
 

bfdd

Lifer
Feb 3, 2007
13,312
1
0
Been gone awhile, Apoppin which card did you like better? I looked at the results pretty even imo could go either way.
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Originally posted by: apoppin
http://www.totalvideogames.com...dsec=download&did=7166

this SC demo? ... it is 1.1GB :p
--is there a built-in bench?

if NOT, then we need to find or create a custom time-demo that we both could share

-BtW - this is a 10 overnight D/L on 56K for me ... and my D/L manager better get it right
What's a 'booster PS'? I have pretty much the same one as you and it's been fine for me, although I haven't really loaded it too much yet. I'm a little discouraged to hear that you've managed to max it out already...
i haven't maxed it - it is fine for oc'ing a single HD2900xt; maybe - just maybe for stock 2900xt Xfire ... but i am missing the extra cables and connectors ...
--a booster PS provides dedicated power for the 2nd GPU and over-clocking options
How do you fit a booster PS into your case?

I can mail you a DVD with the demos if you need me to (yes, the SC one is over a gig). I think even Canada Post takes less than 10 nights. :Q

There are no built-in benches, but I'm pretty sure you can save replays. Let me make a replay; I'm good with GDI and can PWN the CPU bots on 'Brutal'. :D

As for SC, I suck, but maybe someone can help us make a good 'replay' to benchmark with. We can just use FRAPS, no?
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: SickBeast
Originally posted by: apoppin
http://www.totalvideogames.com...dsec=download&did=7166

this SC demo? ... it is 1.1GB :p
--is there a built-in bench?

if NOT, then we need to find or create a custom time-demo that we both could share

-BtW - this is a 10 overnight D/L on 56K for me ... and my D/L manager better get it right
What's a 'booster PS'? I have pretty much the same one as you and it's been fine for me, although I haven't really loaded it too much yet. I'm a little discouraged to hear that you've managed to max it out already...
i haven't maxed it - it is fine for oc'ing a single HD2900xt; maybe - just maybe for stock 2900xt Xfire ... but i am missing the extra cables and connectors ...
--a booster PS provides dedicated power for the 2nd GPU and over-clocking options
How do you fit a booster PS into your case?

I can mail you a DVD with the demos if you need me to (yes, the SC one is over a gig). I think even Canada Post takes less than 10 nights. :Q

There are no built-in benches, but I'm pretty sure you can save replays. Let me make a replay; I'm good with GDI and can PWN the CPU bots on 'Brutal'. :D

As for SC, I suck, but maybe someone can help us make a good 'replay' to benchmark with. We can just use FRAPS, no?

==========
http://www.bjorn3d.com/read.php?cID=1099&pageID=3415

according to this, SC full game has a built-in benchmark

http://www.bjorn3d.com/read.php?cID=1099&pageID=3415

--and also according to them, the GPU is less important then the CPU
Supreme Commander

- 4xAA/16xAF
- Everything set to the highest quality setting
- Built in benchmark used. The benchmark runs a map with 4 opponents that blast each other. It runs for about 7.5 minutes.

***

The HD2900XT performs pretty similar to the GeForce 8800GTS, beating it with a few frames at the higher resolutions. Looking at a graph of the actual fps during the whole benchmark, recorded with fraps, it is obvious that there really is not much difference between the different cards in this benchmark. The game relies heavily on the CPU and it is just in a few instances that you can see a definitive difference between the cards.

yes, we need to "record" a scene and make it into a time demo ... using FRAPS - or find one on the 'net that is made for the demo.

However, it would be SO MUCH nicer to get the "built-in" benchmark of the SC full game ... as they keep patching the game ... i doubt they bother to patch the demo

i have all my other benches up-to-date [HL's LC/FEAR/STALKER/PREY/CoJ [dx10]/LP [dx10] and FC]

=========================
Originally posted by: bfdd
Been gone awhile, Apoppin which card did you like better? I looked at the results pretty even imo could go either way.
i like my HD2900xt for $320 better
-the 8800GTS-640M OC would have cost me over $75 more

the GTS seemed a slightly stronger performer at 16x12 everything maxed plus 4xAA/16xAF turning on the AA slowed down the XT more then the GTS ... at 14x9 [what i play] they are equal

since i have a CrossFire MB, it makes sense also for me to keep the 2900xt ... power is not an issue [for me] and the 2900xt is NOT a noisy card ... the GTS is just a very quiet one.

toss a coin


 

bfdd

Lifer
Feb 3, 2007
13,312
1
0
Ya they looked about even, except at the higher settings. Definitely a good reason to keep the hd2900xt over the GTS if you have a Xfire mobo, I wouldn't ditch my 8800 gtx because of my sli mobo atm. We'll see what happens in the future though. I want to pick up a Penryn when they come out and ddr3 hits 1600mhz or faster.
 

Stumps

Diamond Member
Jun 18, 2001
7,125
0
0
I finally managed to get my hands on a Sapphire 2900XT...which I "borrowed" from work...so far I haven't noticed any real performance difference between the 2900XT and my GTS 640...my GTS still seems to be a tiny bit faster but it not really worth going in to any great detail about.

I do still prefer ATI's vista drivers over Nvidia's...but other than that, both cards are pretty much equal IMHO.

My take so far would be purchase what ever is cheapest...you won't be dissappointed either way...but beware the 2900XT (sapphire's model) Fan is pretty noisy, compared to my Gigabyte 8800GTS, but it still is quieter than some other video cards out there...
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
so letsee ... i think all of OUR benches and conclusions pretty much agree
- with little variations probably due to preferences of game settings, CPU speed, resolution used, driver state and probably even the MB and chipset used ... with even a touch of personal bias thrown in ... and yet ...
... such *uniformity* at ATF Video ... previously unimagined. :p

- also so unlike certain web sites exaggerated conclusions and comparisons.
*cough*

i think i like our 'in house' method - so far ... lets keep it going!
 

Stumps

Diamond Member
Jun 18, 2001
7,125
0
0
Originally posted by: apoppin
so letsee ... i think all of OUR benches and conclusions pretty much agree
- with little variations probably due to preferences of game settings, CPU speed, resolution used, driver state and probably even the MB and chipset used ... with even a touch of personal bias thrown in ... and yet ...
... such *uniformity* at ATF Video ... previously unimagined. :p

- also so unlike certain web sites exaggerated conclusions and comparisons.
*cough*

i think i like our 'in house' method - so far ... lets keep it going!

I agree...

I didn't see any point to adding some benchmarks with my results, simply because they are almost identical to everybody elses who posted them in this thread.