No more AntiDetector on future nVidia drivers?

[WaltC]

I think that you're on the right track, but you're not going far enough. The real problem is that it requires more than just competency to produce a proper analysis of image quality - ie. one with a real understanding of what is going on. To be done properly I believe that it requires experience of 2D and 3D graphics (in a technical sense), and a considerable level of understanding of what is going on 'under the hood' so to speak. In addition to this it then requires much longer and more detailed analysis than that normally allowed for by the amount of time available when constructing a review.

This is a level of investigation that is difficult for web reviewers to achieve. I don't think that there are many review websites that possess people with the appropriate skills and level of knowledge. From the skills that they have it is inevitable that many of these people will tend to work within the industry anyway, and therefore may not be available.

That said, image quality analysis in reviewing has always been the poor relation compared to the task of just churning out benchmark numbers. It is past time that it took its place as a first class citizen, and I think that this is a task that websites are going to have to have to give very careful consideration to in the future if they really want to do benchmarking 'right'.

- Andy.

Aside from making the minor point that I don't think it's really possible to be competent in a field without understanding the issues relevant to it, I agree with everything you've said.

I think what is being overlooked here, though, is the commercial dynamic to "information" of this sort. These companies exist to make products to sell at a profit. The information they circulate about their products is ancillary to that goal. If possible, companies seek to influence how their products are portrayed in the media, and have their own Public Relations departments tasked with such oversight. A such, companies "spin" their product information so as to best facilitate the primary goal of selling as many of those products as they can produce.

It is naive to think that Internet hardware review sites are not influenced by the companies whose products they review. What is different among the hardware review sites is the degree to which they allow their opinions to be shaped by these corporations. There simply can be no uniformity among websites in this regard, just as there is no uniformity among individual human beings in the sense that they are not carbon copies of each other.

So there will inevitably and always be differences of opinion among hardware review sites. This is healthy in my view. After all, it is possible to know how "good" one web site is compared to others only when we can see how "bad" some of the other web sites can be... The alternative is some central authority, like the state, proscribing what can be said and what cannot be said among media outlets.

IE, an "informed" consumer is a forearmed consumer. There is no "easy" method by which one becomes "informed" in a free market economy--you can't go to sleep one night listening to subliminal tapes and awaken the next morning an "informed" consumer, just as one cannot be born on Tuesday and enter the job market on Wednesday...

Yes, I suppose it would be nice if every website always printed the "correct" information the first time and everyone "understood" everything perfectly before they opined. But that, in my view, is not life on planet Earth as I have come to know it...

 

[nelg]

To be done properly I believe that it requires experience of 2D and 3D graphics (in a technical sense), and a considerable level of understanding of what is going on 'under the hood' so to speak. In addition to this it then requires much longer and more detailed analysis than that normally allowed for by the amount of time available when constructing a review.

This is a level of investigation that is difficult for web reviewers to achieve. I don't think that there are many review websites that possess people with the appropriate skills and level of knowledge. From the skills that they have it is inevitable that many of these people will tend to work within the industry anyway, and therefore may not be available.

That said, image quality analysis in reviewing has always been the poor relation compared to the task of just churning out benchmark numbers. It is past time that it took its place as a first class citizen, and I think that this is a task that websites are going to have to have to give very careful consideration to in the future if they really want to do benchmarking 'right'.

- Andy.

Maybe we should just give The ATI guys (Andy, Eric, Terry, etc.) Nvidia cards to review and vice versa. This would allow people who have the level of understanding necessary to cut through the crap to give us the truth.

 

[Himself]

LOL That might be crazy enough to actually work.

 

[euan]

If you can't hack the driver, hack the benchmark.

At least that way one side can't use the excuse of the hacked drivers removing "valid optimisations".

Problem solved.