FutureMark and NVidia in bed

  • Thread starter Thread starter Tony DiMarzio
  • Start date Start date
bullshit, if so why does the 59xx get horrible mark05 scores. Even a 9600
gets a higher score than the ultra. I think aquamark is more useful in
telling you how well your system (graphics/cpu/memory) compares to others in
systems than mark05. I believe mark05 is more like a benchmark of specific
directx features than a benchmark of the processing power of a graphics card
 
Did you read the article at the link? FutureMark has included proprietary,
NVidia specific, accelerations in their '05 benchmarks that have nothing to
do with the DirectX 9.0c specification. What's worse... they've enabled
these NVidia specific optimizations (which can yield 20% performance gains),
but also DECREASE image quality, by default. These actions are in direct
contradiction to FutureMark's claim that they "Would never use
vendor-specific optimizations to render graphics in their benchmarks."

If that wasn't enough to make your stomach turn... they absolutely refused
ATI's request to include 3Dc support in 3DMark05. 3Dc, while not currently
part of the official DirectX specification, has been ratified and will
eventually be part of the DirectX suite. The same can not be said for
NVidia's DST acceleration.

This is not an NVidia bash. This is a FutureMark bash. FutureMark has
succumbed to money and temptation and in the process sacrificed their own
product and brand integrity. Did they really think we wouldn't find out??

Tony
 
Tony said:
This is not an NVidia bash. This is a FutureMark bash. FutureMark has
succumbed to money and temptation and in the process sacrificed their
own product and brand integrity. Did they really think we wouldn't
find out??

Why then are the Radeon cards beating Nvidia cards hands down in 3d
Mark 05 benches?

regards

@ndrew
 
Predator said:
not only that, almost all of the nvidia drivers are not approved.


If you looked at your information correctly you would know that ATI
released a driver specifically for 3d Mark 05. Before this driver
release the X800s and 6800s were benching about the same.

Now all we need to complete the charade is for Nvidia to also release a
cheats driver .. sigh graphics cards companies sucketh in the extreme.
 
CapFusion said:
Shawk said:
Aside from the wrapping I cant get to Hexus.net at all. Anyone else or
just me? Cheers, Shaun

I believed it just you. Try this link instead -
http://tinyurl.com/4fzz5
http://www.hexus.net/content/reviews/review.php?dXJsX3Jldmlld19JRD04NzgmdXJsX3BhZ2U9OQ==
[watch-out for URL wrap]

CapFusion,...

Apparently if you do this
http://www.hexus.net/content/reviews/review.php?dXJsX3Jldmlld19JRD04NzgmdXJsX3BhZ2U9OQ==
It prevents wrapping - I will see when I post this. ;-)
 
Tony DiMarzio said:
Did you read the article at the link? FutureMark has included proprietary,
NVidia specific, accelerations in their '05 benchmarks that have nothing
to
do with the DirectX 9.0c specification. What's worse... they've enabled
these NVidia specific optimizations (which can yield 20% performance
gains),
but also DECREASE image quality, by default. These actions are in direct
contradiction to FutureMark's claim that they "Would never use
vendor-specific optimizations to render graphics in their benchmarks."

If that wasn't enough to make your stomach turn... they absolutely refused
ATI's request to include 3Dc support in 3DMark05. 3Dc, while not currently
part of the official DirectX specification, has been ratified and will
eventually be part of the DirectX suite. The same can not be said for
NVidia's DST acceleration.

This is not an NVidia bash. This is a FutureMark bash. FutureMark has
succumbed to money and temptation and in the process sacrificed their own
product and brand integrity. Did they really think we wouldn't find out??

Tony

Well thats why I never take notice of benchmark software.
When I used to care about benchmarking, I would just record a few demos in
some of the games I was actually playing and use those to gauge the
improvement or otherwise of newer drivers.
 
This is not an NVidia bash. This is a FutureMark bash. FutureMark has
succumbed to money and temptation and in the process sacrificed their own
product and brand integrity. Did they really think we wouldn't find out??

When 3dmark first appeared no one I read ever realised the problem it would
create. Instead of card makers getting their drivers to work better with
real games they have been forced to "tweak" them to get better 3dmark
results. Why? Becuase reviewers and the plebs that read them make such a big
thing of the 3dmark results.
 
When 3dmark first appeared no one I read ever realised the problem it would
create. Instead of card makers getting their drivers to work better with
real games they have been forced to "tweak" them to get better 3dmark
results. Why? Becuase reviewers and the plebs that read them make such a big
thing of the 3dmark results.

Right on......

Isn't it more important that video drivers work properly and be both
forward AND backward- compatible with games and pro-software
than have to satisfy stupid benchmarks-for-nerds.......whose only
thrills seem to be posting benchmark results and complaining
about benchmark compatibility ?? I am far more interested in
seeing shadows properly rendered in Thief 3 ( for example ).
Have Ati got their WHQL drivers right yet on the Thief3 shadows?
They were on their 4th driver since Thief3 released and still
had not got it right when I last bothered to check... Anyway,
Ati cannot possibly do proper regression QC with a 3-4 week
cycle of driver-releases.

Atii (and Nvidia) QC group should post the following definition of
regression-testing in their work-area and insist that all read,
memorize and endeavor to implement it to the best of their
abilities:--

Regression Testing Definition:-
"The selective retesting of a software system that has been modified
to ensure that any bugs have been fixed and <<that no other previously
working functions have failed as a result of the reparations>> and
that newly added features have not created problems with previous
versions of the software. Also referred to as verification testing,
regression testing is initiated after a programmer has attempted to
fix a recognized problem or has added source code to a program that
may have inadvertently introduced errors. It is a quality control
measure to ensure that the newly modified code still complies with its
specified requirements and that unmodified code has not been affected
by the maintenance activity"

( The << and >> have been added by me )

At least nVidia seem to have their official-release driver priorities
straight. I have an extensive set of current and legacy game titles
and pro-video software. nVidia's driver compatibility has been
excellent overall, except for a dark patch early this year when they
were optimizing the run-time compiler for the FX series.( Ignore
anything between 53.03/53.04 and 56.72 )

And WHQL means nothing. My pro-video software crashed on
nV 56.64WHQL -- had to go back to 53.04. Fine on 56.72(WHQL too) --
issued 2 weeks later; somebody in nVidia QC was obviously on-the-ball
with regard to rushing a fixed release out.

John Lewis
 
Back
Top