I'm very impressed by what Gibbo has done for us and that OcUK is offering refunds for those who want them. I will be buying all my PC equipment from OcUK from now on, which I have not done in the past. However, I've just completed a major upgrade. The motherboard and CPU came from Amazon (another retailer who has dealt very fairly with me concerning returns in the past), the graphics card from OcUK.
I'm relieved that the news broke just in time for me to change my pre-order for an OcUK reference 970 to a 980. If I'd already received and installed the 970, I would have kept it as I think it would have been fine for playing at 1920x1200. However, I upgraded to a 980 because I didn't want to take the chance that I'd regret having purchased the 970 in 12 months' time. I think a lot of console ports will need 4GB of VRAM. I looked long and hard at the AMD 290X, but concluded that both the 970 and 980 would still suit my needs better. While the 290X is faster than a 970 at 4K, it's slower than a 970 at 1080p. It grated to spend more money with NVIDIA, but I felt I'd be cutting my nose off to spite my face if I bought a product that didn't fit my needs as well, just because it wasn't from NVIDIA.
I'm disgusted by the way that NVIDIA has behaved. I don't think they have a leg to stand on legally. It's clear to me that NVIDIA has engaged in false advertising. I suspect they may try arguing that they never advertised that the 970 had the same memory architecture as the 980, and that misinformation had been spread by third party reviews. For example, I just checked the review in Custom PC from when the 970 and 980 were launched. It reads, "Elsewhere, the full 4GB of GDDR5 memory again runs at 7GHz (effective) across the same 256-bit interface, and all 64 ROPs are intact too". We now know this statement to be false. However, Custom PC would have based their review on the reviewer's guide supplied by NVIDIA. Jonah Alben, NVIDIA's SVP of GPU Engineering has admitted that there was an error in the reviewer’s guide and a misunderstanding between the engineering team and the technical PR team on how the architecture itself functioned.