nVidia GTX 960 2GB - what should I expect?

[OnlyforDCS]

My point is that it's not an issue of performance. It's an issue of mislabeling and consumer trust. Nvidia has lost some trust among it's consumers because they decided to lablel a card falsely that it has 4GB of VRAM, when in effect it really has 3.5 GB of high performance VRAM (the rest is disabled with drivers in a high vram usage scenario).

 

If you really want a 4GB card then by all means, look somewhere else, but my original point still stands, you wont find a better performing card (even one with the full 4GB of VRAM) in the same price range as the GTX 970.

[blackbelter]

My point is that it's not an issue of performance. It's an issue of mislabeling and consumer trust. Nvidia has lost some trust among it's consumers because they decided to lablel a card falsely that it has 4GB of VRAM, when in effect it really has 3.5 GB of high performance VRAM (the rest is disabled with drivers in a high vram usage scenario).

 

If you really want a 4GB card then by all means, look somewhere else, but my original point still stands, you wont find a better performing card (even one with the full 4GB of VRAM) in the same price range as the GTX 970.

 

When did I say 970 is not a good card as it is? Consider someone who bought it because its "4GB of vram" is somewhat future proof. Is it good for them? And as far as I have read, the low performance vram does indeed negatively impact performance. That's how it was discovered in the first place. I have not read anywhere that this part of the vram will be disabled completely.

 

And next time, before you point your finger at others and accuse them of being "disinformation" and saying "rubbish", make sure you are standing on a more solid ground. Changing posture when confronted and pretending that you were saying something else does not make your argument more convincing, nor yourself more polite.

 

I am done with this thread. Bye.

[OnlyforDCS]

Wow. Sticking to your guns no matter what? A fine and admirable quality except in this case your point was and is still wrong. You are the one who claimed that high VRAM usage applications somehow run much slower because of the slower ram, well that's how can I put this delicately, 'rubbish'. They dont because Nvidia's drivers wont let them even 'see' or access the memory in question. It's not 'magic' as you so eloquently put it, its basic I/O driver design.