x1900xt scored 9004 marks stock, 9688 overclocked

Quote:

---

Originally Posted by hipro5

Don't worry........On Monday/Tuesday I'll get an X1900XTX to test.......I already have an 7800GTX 512MB and an X1800XT in my hands......I'll compare all three of them together with the EXACT SAME config........;)

---

sounds nice to me !;)
please don't rape the orb to much for now :D

Quote:

---

Originally Posted by Willis

sounds nice to me !;)
please don't rape the orb to much for now :D

---

No I'll rape only 2k1 with the X1800XT.......and IF "they" leave me, with the X1900XTX as well.......:D ;)

Quote:

---

Originally Posted by Willis

This is what i call fanboy talk :)
It's not X1900XT vs 7800 it's vs 7900.
And when it's there i think allot in here including you will be silenced.

---

you talk as fan boy too

at this moment the competition is between cards with similar price
if the X1900XTX cost as the 7800gtx-512 :fact: , then this is the comparison

when the 7900 get out, then we'll se if this will cost more or same as X1900XTX, if this will happen, we'll make some comparison

whats the word im looking for without being to cruel.....o yes Nvidiot

Quote:

---

Originally Posted by Willis

please share a link with me :)

---

why don't brother

http://www.google.it/search?hl=it&q=...n+Google&meta=

Yes, X1900XT is in 3dmark05 slow like X1800XT because 3dmark05 is Vertex shader based test and X19xxx has 16 shaders like x18..., my score: X2@2800+ X1900XT@stock - 9800points, overclocked@700/800 = 10500points.

Quote:

---

Originally Posted by Willis

aimed at who?

---

the thread starter
Reason being he didn't do his homework so its an open mouth insert foot situation. I'm not saying any card is faster till i get one in my hands. Atm nvidia is tearing up 3d06. Thats all fine and dandy. The problem is the score that was posted is way out of proportion.

Quote:

---

Originally Posted by Willis

i meant it's about comparing the same cards from teh same genaration :)

i

---

this is sensless

when you pay for a card 649$, what you can get is

a) 7800gtx-512
b) X1900XTX

is hard to understand how it work? :stick:

if you like to be a fanboy, this change the whole thing, then what matter is 'generation'

Quote:

---

Originally Posted by Willis

a picture from this would be nice :)

---

Will be.. lately.

Quote:

---

Originally Posted by Overmelted

a truly 'non-existant' X1900XL?:slapass:


there's only two model of X1900 family
the XT (620 MHz) and the XTX (650MHz):fact:

this is the XT, not XL:nono:

---

Come on......Gimme a break......Dothan @ 2600MHz and 200MHz fsb 1:1 rams, the 7800GTX 512MB gives me 9999 marks 2k5, the X1800XT with the same config, gives me 9254 marks 2k5.......So there is NO WAY the X1900XT give ONLY 9000 marks on 2k5.......:)

Quote:

---

Originally Posted by OBR

Yes, X1900XT is in 3dmark05 slow like X1800XT because 3dmark05 is Vertex shader based test and X19xxx has 16 shaders like x18..., my score: X2@2800+ X1900XT@stock - 9800points, overclocked@700/800 = 10500points.

---

3dm06 is useless, because with some cards it go with partial precision, with other go with full 32 bit internal precision

we have to use GAMES in order to make a fair comparison

Quote:

---

Originally Posted by OBR

Yes, X1900XT is in 3dmark05 slow like X1800XT because 3dmark05 is Vertex shader based test and X19xxx has 16 shaders like x18..., my score: X2@2800+ X1900XT@stock - 9800points, overclocked@700/800 = 10500points.

---

Thats not OCed well at least not anything like some of us Xtreme Oc'ers do. That would be my 24/7 speed........how bout you try getting it up to say 816/981 lol then its oced and it probably has more headroom then that.......

Quote:

---

Originally Posted by Overmelted

3dm06 is useless, because with some cards it go with partial precision, with other go with full 32 bit internal precision

we have to use GAMES in order to make a fair comparison

---

3d mark 06 is not useless....... me personally i think ive beat just about all 7800 gtx 256 cards and give the 512mb card a run for its money.

Quote:

---

Originally Posted by Willis

I wil never pay 649$ for a card ;) ,
And i believe this thread was meant about discussing the performance not the price :)

---

you are going to make me crazy :(

when you make a fair comparison between two cards, you have to compare two card of SIMILAR PRICE

or you are some of those who compares 7800gtx-512 with 1800xl, X1300 with X850, and so on??

what is as hard to understand that two cards are RIVALS when the are rival in price????
:slapass:

Quote:

---

Originally Posted by sabrewolf732

Dude, THIS IS 3DMARK! I thought you officially retired from 3dmark btw? Why do you put so much belief into it. Results in fear show around 50% improvement over a 512GTX, that's nothing to sneeze at. I highly doubt even the 7900gtx will beat the 1900xtx in fear, it has a 33% increase in pipes and 36% increase in clock.

---

I retired from benching but I still look at 3dmark to compare gpu power. I can see that the x1900xt scored in the 9000s. FEAR is just one game and its very very shader intensive, something ati is always good at, even the x1800xt wins that one. Lets see about the 20 other games. 33% increase? uh the 7900gtx has 100% increase in pipes and its clocked higher too!



"Don't worry........On Monday/Tuesday I'll get an X1900XTX to test.......I already have an 7800GTX 512MB and an X1800XT in my hands......I'll compare all three of them together with the EXACT SAME config........I think that this will be fair enough......."


you are my hero! :woot: :toast: be sure to test in all four versons of 3dmark as well as whatever games you have! Keep the cpu, ram, etc constant and the video cards stock then overclocked.


"It's not X1900XT vs 7800 it's vs 7900."


Then why did ati release the x1900xt now and with only 16 pipes?



"you talk as fan boy too

at this moment the competition is between cards with similar price
if the X1900XTX cost as the 7800gtx-512 , then this is the comparison

when the 7900 get out, then we'll se if this will cost more or same as X1900XTX, if this will happen, we'll make some comparison"


My point exactly. ATI has priced their x1900xt to compete with 7800gtx 512mb....


"Dothan @ 2600MHz and 200MHz fsb 1:1 rams, the 7800GTX 512MB gives me 9999 marks 2k5, the X1800XT with the same config, gives me 9254 marks 2k5.......So there is NO WAY the X1900XT give ONLY 9000 marks on 2k5......."


Hmmm id probably say youd get just over 10k with the x1900xt. :confused:

Without a CPU-Z shot/card clocks displayed those screen shots are useless. Even if they are valid it's obvious that system was setup by someone who doesn't have much skill.

The VR-Zone tests I trust.

there is to much you are a fanboy -no you are a fanboy,i've red this word more in 2 days about nvidia ati discussions then in my whole life:D:D!!we don't want a nvidia section and an ati section right? why not be more calm... we have just to wait and everything will be clearer! anyway i personally think that 7900 will have an higher performance increase over 7800 then 1900 has on 1800...to see if my impression are true i'll wait:D:D

Quote:

---

Originally Posted by OBR

Yes, X1900XT is in 3dmark05 slow like X1800XT because 3dmark05 is Vertex shader based test and X19xxx has 16 shaders like x18..., my score: X2@2800+ X1900XT@stock - 9800points, overclocked@700/800 = 10500points.

---

Umm this sounds wrong. X1800/X1900 are both more powerful in vertex shading than the 7800 series. GTXs and X18/19s all have 8 vertex shaders, so difference in vertex performance should come down to gpu clock frequency, which is currently higher on ATI.

Furthermore X1900s don't have 16 "shaders", rather 48 shading units (ALUs). The only 16 parts about them are the TMUs and the ROPs.

Quote:

---

Originally Posted by chew*

3d mark 06 is not useless....... me personally i think ive beat just about all 7800 gtx 256 cards and give the 512mb card a run for its money.

---

this don't change anything
nividia run the test with internal partial precision (16/24) as old ati (x800 and so on), the 1X00 family run all the tests with full 32 bit precision
if you think that this is a fair test, ok

then look at the benchmarks, the only game that reflect the perf ratio between the cards is Doom3

Quote:

---

Originally Posted by Willis

i still see FX5950Ultra hovering around for 649$
i don't see you comparing them?

---

If you can't see a difference, get a brain
I want to stop this useless discussion with a person that don't want to discuss and say this kind of bs
:nono:

im sick of all this rumor BS, lets wait until the real owners bench

Welcome Manic One to XS :)

For the rest of you going back and forth here...for god sakes, wait 72 hours.

Perkam

Quote:

---

Originally Posted by perkam

Welcome Manic One to XS :)

For the rest of you going back and forth here...for god sakes, wait 72 hours.

Perkam

---

I think that X1900 will not score much more than X1800
VR zone is showing a lot of benchs already..

1600x1200 no aa 71 fps
1600x1200 aa4x af8x 49 fps
1600x1200 aa8x af16x 48 fps
1600x1200 SS ON 38 fps
____________________________________________
http://resources.vr-zone.com/newspic...1200-SSoff.jpg
http://resources.vr-zone.com/newspic...x1200-SSon.jpg
http://resources.vr-zone.com/newspic...0-4xAA8xAF.jpg
http://resources.vr-zone.com/newspic...xAA16xHQAF.jpg
http://www.vrforums.com/showthread.php?t=53164&page=8

Quote:

---

Originally Posted by Overmelted

this don't change anything
nividia run the test with internal partial precision (16/24) as old ati (x800 and so on), the 1X00 family run all the tests with full 32 bit precision
if you think that this is a fair test, ok

then look at the benchmarks, the only game that reflect the perf ratio between the cards is Doom3

---

where did you get these info?
nvda never support 24bit precision............

Quote:

---

Originally Posted by Victorshen

where did you get these info?
nvda never support 24bit precision............

---

errata corrige
I mean '16/32'