AMD 7nm GPU News and Rumours 2018/2019

[Gerard]

Not really interested in Gamers Nexus, I am still laughing at them when they made a big thing about putting an AIO on a Titan V and tweaked their setup to set some benching records. It took me one go to beat their scores with a standard air cooled Titan V because they did not know enough about tweaking the CPU memory.

You should not believe everything some of these tech sites claim, it is far better to test for yourself..

So you still are of the conclusion that hbm using lower power is a myth despite the fact nobody (besides you) claims otherwise?

 

[Kaapstad]

I don't think AMD ever considered using HBM because of its power saving abilities.

I think AMD decided to use HBM because at the time it was the cutting edge option for its other properties like high bandwidth.

AMD did something positive when they included HBM in the design of the Fury cards but unfortunately because of factors outside their control it did not work out.

 

[4K8KW10]

So you still are of the conclusion that hbm using lower power is a myth despite the fact nobody (besides you) claims otherwise?

AMD in their slides say 3 times higher performance per watt than GDDR5.

 

[Kaapstad]

So you still are of the conclusion that hbm using lower power is a myth despite the fact nobody (besides you) claims otherwise?

I think it does use less power but not as much as people seem to think and not enough to justify its use purely on those grounds.

 

[Rroff]

AMD in their slides say 3 times higher performance per watt than GDDR5.

Potentially it can but I don't think anyone has implemented it so far to take advantage of that headline figure - also harder to utilise that advantage for gaming purposes as the way HBM works with best efficiency is more compute type workloads which currently don't make up the bulk of gaming processing.

 

[Kaapstad]

AMD in their slides say 3 times higher performance per watt than GDDR5.

I remember that slide well as I have used it in other threads.

Do you believe everything that manufacturers claim, next you will be telling me how fantastic NVidia's Turing card launch was.

 

[Gerard]

I think it does use less power but not as much as people seem to think and not enough to justify its use purely on those grounds.

So why do you call it a "myth" if it's fact? Gddr5 per module was meant to be 1.5v Vs hbm 1.3 v. How this translates to gddr6 I don't know. But overall it's less power so a legitimate claim nomatter what.

 

[Kaapstad]

So why do you call it a "myth" if it's fact? Gddr5 per module was meant to be 1.5v Vs hbm 1.3 v. How this translates to gddr6 I don't know. But overall it's less power so a legitimate claim nomatter what.

Because the figures are tiny compared to what some people think HBM can do.

 

[Gerard]

Because the figures are tiny compared to what some people think HBM can do.

It's still not a myth.

 

[Kaapstad]

It's still not a myth.

When people start quoting large figures it is definitely a myth.

 

[Gerard]

When people start quoting large figures it is definitely a myth.

That's up to them, people exaggerating on tech forums is nothing new.

 

[doody]

someone that knows what he's talking about https://www.youtube.com/watch?v=MG-onUm__c8;)