How/why did Amd/ATI fall behind Nvidia?

[happy medium]

Have threads like these accomplished anything over the years other than to corral those who participate?

I think this is a excellent point. Take a close look at who is derailing the thread, totally against the thread, and usually you will have your paid participants.

 

[happy medium]

I'll bet you have no idea who is (if any) and who isn't.

I do now.

 

[Blitzvogel]

Isn't Maxwell's efficiency and success based on having a large amount of L2 cache? More cache means less dependency on external VRAM, meaning reduced board power and board complexity, and therefore cost........yes? If that is the case, it seems like a very logical thing for AMD to "bolt on" more of into GCN, even more so than moving the entire current line of GPUs to HBM.

Edit: Not just cache, but color compression, and being a smaller die to boost GPU clocks.

 

[happy medium]

If that is the case, it seems like a very logical thing for AMD to "bolt on" more of into GCN

They will, but they are just a little behind.

 

[NTMBK]

2 dies, even if interposer means you need a lot of highspeed interconnect between the 2. Thats both power and die size increase.

Meh, a pretty small increase. Pushing signal over an interposer takes hardly any more power than sending it over a long wire on the same die. Trade off is that you get smaller dies with better yields.

 

[Keysplayr]

n/m

 

[blastingcap]

n/m

 

[Headfoot]

Aren't you being a hypocrite?

Directed to no particular poster here, but hypocrisy is the default in internet forums

 

[Keysplayr]

n/m

 

[stahlhart]

Thread closed for moderator review.
-- stahlhart