ATi R600

[Masster]

No tak ti co doufali, ze "spatny" vykon v DX9 nic neznamena a ze to DX10 zachrani, tak timhle asi nebudou nadseni.



http://we.pcinlife.com/thread-763485-1-1.html

[xjohny]

Me by zajimalo, co je mysleno timhle
http://ati.amd.com/online/getall3/

[Pavlis]

Me by zajimalo, co je mysleno timhle

rek bych Kuma+R600+RD790

[Masster]

Fantastic specs don't always add up to a lead in real-world applications, but ATI has been able to convert its technological advantages into some lethal framerates. It didn't beat the 640MB version of the 8800 GTS in all of our game tests, but it did come awfully close (see chart, below). The advantage is greatest at a resolution of 1,024x768, which suggests that the less detail you want to stack on, the better this card will prove.

http://www.pcadvisor.co.uk/reviews/i...m?reviewid=834

[Q]

s kazdyma novyma ovladacema stoupa vykon v desitkach procent...pockejme si...

[THX]

Ati malo prve vzorky hotove snad uz pred pol rokom. Architekturu velmi nemenili, takze uz snad pol roka mohli vyvijat a testovat drivery - teda, tie aktualne drivery by uz mohli byt celkom odladene. Minimalne isty cas na odladenie ziskali koli problemom s vyrobou, ak by mohli kartu by vydali urcite skor a na to museli mat pripraveny nejaky driver.

Takze povedal by som, ze sa vykon samozrejme este zvysi, ale nejake velmi radikalne narasty vykonu (a casto sa opakujuce po kazdej novej verzii driverov) by som necakal, mozno spociatku prvych par vydani drivera (2-3) ale potom uz asi velmi nie.

[dexterav]

tých 26 fps by potrebovalo riadny booost ...
ale zatiaľ sú to stále fámy a v podstate iba fantazírovanie...
počkajme na ofiko a potom sa uvidí, ale veriť že všetko bude makať na 100% je trochu naivné...

[Pavlis]

Takze povedal by som, ze sa vykon samozrejme este zvysi, ale nejake velmi radikalne narasty vykonu (a casto sa opakujuce po kazdej novej verzii driverov) by som necakal, mozno spociatku prvych par vydani drivera (2-3) ale potom uz asi velmi nie.

mno co ted prosakuje o ruznejch verzich ovladacu tak ty narusty jsou docela velke, uvidime co ukaze finalni verze. jinak uz x19xx rada ukazala jak jeji vykon roste s case, takze bych tuto moznost neodmital, osobne bych rek ze tady je prostor jeste vetsi

[Pavlis]

No tak ti co doufali, ze "spatny" vykon v DX9 nic neznamena a ze to DX10 zachrani, tak timhle asi nebudou nadseni.

tady je to zase naopak

http://forum.beyond3d.com/showpost.p...postcount=5253

[no-X]

No tak ti co doufali, ze "spatny" vykon v DX9 nic neznamena a ze to DX10 zachrani, tak timhle asi nebudou nadseni.

http://we.pcinlife.com/thread-763485-1-1.html

hmm, ale ještě k tomu chybí dost podstatná poznámka:

Note that this implementation is relatively unoptimized; it's more of a "proof of concept". In particular, the shadow split locations could be chosen a lot more sensibly, and even some basic frustum culling would greatly improve the performance of rendering the different shadow map splits.

[dexterav]

najhoršie je že sú prakticky iba 2 druhy testov...
prepadák a šupa

[Masster]

8.37


1600x1200 FA@16x, FAA@4x SS
2900XT/8800GTS:
min:17/20
med:35/37
max:64/70

1920x1200 FA@16x, FAA@4x SS
2900XT/8800GTS:
min:16/16
med:30/32
max:55/57


2900XT - 8.37- 8800GTS - 158.19 -

1600x1200 FA@16x, FAA@4x no SShadows,
2900XT/8800GTS:
min:20/30
med:59/58
max:124/130

1920x1200 FA@16x, FAA@4x no SShadows,
2900XT/8800GTS:
min:19/27
med:51/51
max:111/104


STALKER 1600x1200 all maximum, AFMax noAA - C2 3200

Timedemo1: Av: 37,55, Min: 19,79, Max: 123,84, Mid: 39,52
Timedemo2: Av: 60,14, Min: 37,60, Max: 291,21, Mid: 65,48

8.36

Timedemo1: Average: 33,12, Min: 17,51, Max: 88,29, Mid: 35,51
Timedemo2: Average: 50,50, Min: 31,28, Max: 282,81, Mid: 56,67

A dalsi

Driver: 8.37-4, CPU: E6700, from europes biggest PC-Magazine ct

Oblivion, HDR on

1280 X 1024 noAA/noAF

2900 XT 49 fps
88 GTS 48 fps
88 GTX 48 fps

1280 X 1024 AA 4X / noAF

2900 XT 39 fps
88 GTS 46 fps
88 GTX 48 fps

1280 x1024 AA 8X / noAF

HD 2900 XT 17 fps
88 GTS 28 fps
88 GTX 39 fps


PREY

1600 x 1200 AA 8X / AF 16X

HD 2900 XT 43 fps
88 GTS 37 fps
88 GTX 50 fps

[no-X]

8.37...

Výsledky z 8.37 jsou už poněkud outdated, když už pár dní existují 8.38:

8.37



8.38

[dexterav]

tak to je pekné ...

[nex]

Tohle vypada lepe

Windows Vista Direct X10 SDK

PIPEGS

2900XT 159 fps
88GTS 34 fps
88GTX 63 fps


CubemapGS, Car, Instancing

2900XT 23 fps
88GTS 9 fps
88GTX 11 fps

Cubemap, Car, Instancing

2900XT 18 fps
88GTS 10 fps
88GTX 11 fps

this seems to be confirming my theory that if R600 is faster anywhere, it will be in geometry shading. Granted G80 still has a lot of space to improve in that area: a recent test that I did indicated that even having *any* geometry shader present (I used an "identity" geometry shader that just passed through the triangles) cut the performance of the shader in *half*.

It will certainly be interested to see... in any case I think the Cubemap examples are a bit contrived to begin with because arguably separately rendering the faces with a good frustum culling algorithm is going to be just as fast or faster except in the most degenerate cases.

What I really want to see is how well R600 runs cascades...

[Masster]

Výsledky z 8.37 jsou už poněkud outdated, když už pár dní existují 8.38

Ano, ale pokud vim, tak recenzenti dostali s kartou 8.36, pak jim teprve byla zaslana verze 8.37 a pochybuju, ze nekteri budou predelavat svoje mereni 3x, nehlede na to, ze z 8.38 byl ukazan jen screen z 3DMarku, ktery nemusi co se tyce vykonu ve hrach vubec nic znamenat.

[Sharker]

najhoršie je že sú prakticky iba 2 druhy testov...
prepadák a šupa

Tak si vyber ..... poradim ti .. podla infa od ludi co tu kartu UZ maju tak sa mame na co tesit .... vykonovy prepadak to nieje ..

[dexterav]

ja vyberať nepotrebujem...
kto ju má?