Geforce 8600/8500/8400/8300

viskonde disse:
isso esta bastante limitado pelo CPU...
nao ha testes por ai com CPU's que acompanham a grafica?
Clicar para expandir...

Sempre podes ver o SM2 e o HDR/SM3 score do 3dmark 2006 que em pouco ou nada é influenciado pelo CPU.

É interessante que em relação a um X1950pro a 8600GTS vence por larguissima margem no SM2.0 score, mas perde no SM3/HDR score, quando deveria ser ao contrario. Será um mau prenuncio para a performance em DX_10 (SM4.0)?

Em todo o caso a competidora da X1950pro é a 8600GT e nao a 8600GTS. A competidora da GTS é a X1950XT (250 euros). Isto so para comparação de scores, pq depois é preciso comparar com as RV630.

EDIT: Essas imagens devem ser fake por causa da discrepância SM2.0 e SM3/hdr score :wvsore:
 
Última edição:
Eu ainda não consigo acompanhar o poder dessas gráficas através do 3dmark2003... Estou tão habituado ao 2001 que não sei avaliar o real poder delas...

Alguém tem alguma conta de cabeça para conversão de pontos do 2003 -> 2001??
 
SwimmerBoy disse:
...

Alguém tem alguma conta de cabeça para conversão de pontos do 2003 -> 2001??
Clicar para expandir...

Penso que isso seja mt dificil se ñ impossivel.

São bench's mt diferentes, um avalia o sistema como um todo (2001) enquanto outro foca-se mais na parte gráfica (2003), existem mts variaveis. Pessoalmente tb prefiro o 2001 por analisar o sistema todo.

Fikem bem!
 
Pois, eu sei que é complicado devido ao 2003 deixar de contar com o CPU...

Mas por muitos valores diferentes que o 2001 possam ter por se ter CPUs e boards diferentes, consigo ter uma noção muito melhor do desempenho da gráfica associada ao sistema!

Também pode ser por nunca ter feito muitos benchs em outros que não o 2001...
 
8600GTS/GT benchmarked

Not CeBIT 007 Nice scores, one big problem

I WAS LUCKY get some quality time with a 8600GTS and 8600GT recently, and happened upon a few numbers you might be interested in. All numbers were measured on an Intel X6800 machine running the latest 3DMark.

The raw scores are 5500 3DMark06 for the GTS and 4800 for the GT, blurred slightly to protect the guilty. That is the good part. The bad part is when you start using heavy textures, performance drops off notably, and I mean notably. Think cliff.

This is most likely because of two things, drivers or bus width. NV drivers are still pretty badly broken, and upcoming cards are probably less of a priority than getting the ones they released almost 6 months ago functional.

The other thing it could be is the narrow memory bus simply choking on all the data. If this is the case, don't look for improvements, this card will always be a benchmark special.

There could be a host of other things, from pre-production boards (doubtful in this case), to sunspots, but the performance drop is quite real. Keep an eye out for this when the boards are finally released on April 17. µ
Clicar para expandir...
http://www.theinquirer.net/default.aspx?article=38346
 
DJ_PAPA disse:
8600GTS/GT benchmarked

Not CeBIT 007 Nice scores, one big problem


http://www.theinquirer.net/default.aspx?article=38346
Clicar para expandir...

Boy, o Charlie e o The Inquirer devem ter feito um downgrade enorme no Core 2 X6800 que usam para fazer testes com o 3DMark 2006.

Isso, ou estão simplesmente a mentir, e a usar scores falsos de terceiros (supostamente feitos com um Athlon 64 3000+ single-core).
Onde estão os screens dos testes deles ?

Espantoso atingirem 5500 pontos no 3DMarks 2006, quando o tal site chinês deu isto com a CPU fraquinha acima referida:

8600c.jpg



Realmente, as coincidências são tramadas... :lol:
 
blastarr disse:
Boy, o Charlie e o The Inquirer devem ter feito um downgrade enorme no Core 2 X6800 que usam para fazer testes com o 3DMark 2006.

Isso, ou estão simplesmente a mentir, e a usar scores falsos de terceiros (supostamente feitos com um Athlon 64 3000+ single-core).
Onde estão os screens dos testes deles ?

Espantoso atingirem 5500 pontos no 3DMarks 2006, quando o tal site chinês deu isto com a CPU fraquinha acima referida:

8600c.jpg



Realmente, as coincidências são tramadas... :lol:
Clicar para expandir...

Mas esse bench tb nao era fake?
 
São fakes, claro.
Apenas achei engraçado que o Charlie do The Inq dissesse que tinha passado um "quality time" com a placa para uns testes, quando isso não é verdade.
 
We have obtained some information on performances of middle line and entry level GeForce 8: the GeForce 8600 GTS, 8600 GT and 8500 GT. These performances are restricted to 3DMark 06 and have to be put into perspective since the benchmark isn't always representative of performances in games.

With the same platform, if an 8800 GTX would obtain 11900 points, a GeForce 8600 GTS would obtain 7000 points or an identical score to the Radeon X1950 XTX. The GeForce 8600 GT with 5800 points would be slightly faster than the Radeon X1950 Pro and GeForce 7950 GT whereas the GeForce 8500 GT (2800 points) would be a little less performing than the Radeon X1650 Pro.

We noted that, for once, Nvidia has given a higher number to the entry level (based on G86). The card name is GeForce 8500 and not "GeForce 8200". Let's hope that this new approach isn't made to rename old products in GeForce 8xxx or have more entry level products and even less performing GeForce 8…
Clicar para expandir...
http://www.behardware.com/news/8684/cebit-geforce-8600-and-8500-performances.html
 
Ah, então agora já são 7000, em vez de 5500, hein ?
Aquele Core 2 X6800 do The Inquirer devia estar num underclock severo... :004:


BTW, a fonte original desse artigo é este link de Sábado passado, não o que o DJ_Papa deu.
 
Última edição:
Overclocked Geforce 8600 GTS comes later




We learned that at least a few Nvidia partners will first go for non overclocked cards. No one wants the Osborne effects where you have a cool product and announce even better one and cripple the sales of your original part.


The first G84 cards branded as Geforce 8600 GTS will work at 675 MHz core / 2000 MHz memory. The later versions will work at more than 700 MHz for the core and even more aggressive memory speed.


None overclocked comes first and plans to take the market and after than when ATI finally launches RV630 based Radeon X2600 series Nvidia and its partners will answer with even faster clock.


There should be more place for overclocking these chips as they are 80 nanometre. The Nvidia reference Geforce 8600GTS is suppose to launch on the April 17th.


http://www.fudzilla.com/index.php?option=com_content&task=view&id=205&Itemid=1
Clicar para expandir...
cumps
 
Another set of GeForce 8600 GTS & GT Benchmarks Leaked

Blastarr disse:
Ah, então agora já são 7000, em vez de 5500, hein ?
Aquele Core 2 X6800 do The Inquirer devia estar num underclock severo...


BTW, a fonte original desse artigo é este link de Sábado passado, não o que o DJ_Papa deu.
Clicar para expandir...

8600GTS
868501.jpg


cebit2007-gf86all86gts03.jpg

cebit2007-gf86all86gts06.jpg


8600GT
868502.jpg

cebit2007-gf86all86gt03.jpg

cebit2007-gf86all86gt06.jpg


AMD FX-62
NF590 SLI
2GB Kingston DDR2-667
Windows XP

Pretty consistent with VR-Zone benchmarks using Core 2 Extreme X6800 : http://www.vr-zone.com/?i=4775

GeForce 8600 GTS : 5,7xx pts
GeForce 8600 GT : 4,7xx pts
http://www.oc.com.tw/article/0703/readolarticle.asp?id=5772
 
Recently showed 8600 GTS scores are a fake

On last Sunday we reported about PCOnline who were showing the worlds first 8600GTS benchmark figures. A few guys nicknamed 'kemo6600' and 'Gam3Ra' over at the XtremeSystems (scroll down to post 15) point out there is something wrong with these figures and the pictures. Admittedly there is something wrong with the pictures in question. There are 4 of them in the thread at XtremeSystems and all of them have a ACDSee EXIF-tag in it (probably for some cropping reasons) besides one! Coincidentally this one picture is the one that states the 3DMark 06 score. If you look carefully at the provided picture you will see the different sub-category scores at the front center and the right window at the back. Look at the picture at the end of this post where I enframed the parts in red.
Anyway there is another way to bust this picture as a fake even if the numbers weren't manipulated. Check out the 3Dmark2006 Score Calculator by M3kk which calculates the final '06 score based on given SM02, SM03 and CPU scores in the same way the real deal arithmetics behind 3DMark06 are doing it. If you enter SM2.0 3249, SM3.0 2070 and CPU 937 you will get 5212 3DMarks only. Now if you calculate using the other values (SM2.0 2356, SM3.0 N/A, CPU 940) it comes up with 3902 3DMarks. The last picture I added show what happens when you combine the most convincing scores (SM2.0 2837, SM3.0 2070, CPU 940) - the 3DMark2006 Calculator v1.2 now renders 4598 3DMarks which is in fact nowhere close to the stated 5488 3DMarks.
You might wonder why I didn't link to the pictures of the source story. That is because the PCOnline is down at the moment and I would really be astounded if the story is still up by the time they go online again. The other reason is that our own picture-upload engine get's rid of the EXIF tags so our own article doesn't give any evidence.



Source: XtremeSystems
Clicar para expandir...



para quem nao esta recordado os scores FAKES sao aqueles de 5488 pontos com 8600GTS e um A64 no 3dmark 2006

os scores reais sao os que estao calculados no 3dmark calculator nestas imagens





cumps
 
Última edição:
Inicie sessão ou registe-se para poder participar.
Back
Topo