Want to jump ship to amd

Johan45 said:

I'd look for a 290x on the cheap. There isn't much difference between it and the new ones aside from more ram

Click to expand...

ATMINSIDE said:

The only thing that'll perform close to the 980 is a Fury X though.

Click to expand...

ATMINSIDE said:

What DX12 business?
The one game, in alpha stage, that utilizes AMD better than Nvidia?

Click to expand...

EarthDog said:

I'm sorry... I still don't understand the point or if what you are saying even matters for what you are using it for. You are on about things that aren't factually correct or just a factor.

Click to expand...

Pipeline state objects[115] have evolved from Direct3D 11, and the new concise pipeline states mean that the process has been simplified. DirectX 11 offered flexibility in how its states could be altered, to the detriment of performance. Simplifying the process and unifying the pipelines (e.g. pixel shader states) lead to a more streamlined process, significantly reducing the overheads and allowing the graphics card to draw more calls for each frame.

Direct3D 12 also learned from AMD Mantle in command lists and bundles, aiming to ensure the CPU and GPU working together in a more balanced manner.

Click to expand...

ozan said:

Nv puts out inferior hardware(based on the raw power) for the price which "just works" with superior drivers and lots of manipulation of the game market which lets them cripple competition repeatedly which strengthens the "just works" image. Tesselation, physx are the ones in the recent history.

So faster GPU's that use less power with better driver support are inferior? Cool story.

Now i will directly quote the directx wiki. Im sorry i cant go through scientific papers for a simple forum post.

Okay, cool, just because AMD can do something doesn't mean that nVidia can't.

Now we can assume amd cards with larger bandwidths can go nuts on dx11 and communicate to their hearts content. Wrong. With the constricted and limited pipeline the larger bandwidth becomes uncompetitive and amd who relies on less driver tweaks and optimizations has to build larger pipelines for their data to go through in time.

As the dx12 and vulkan emerges, the pipeline is expected to be largely liberated therefore largely reducing the need for pipeline optimizations.

The pipeline optimizations, also known as data compression, allow nVidia to pass more data across a smaller bus than AMD can.
Neither of these companies are limited by the current bus capabilities though, so I don't see how your point is valid in any way, shape, or form.

Better chip is better, that is for sure but amd and nv's production of chips dont variate greatly by the means of nm and manufacturer. It's rather what they put in the chips and what they optimize it for.

You're right, better silicon design is better. And nVidia has the better silicon.
See above "more performance, less power" comment.

I think nv is all in the dx11 train still and has research to do before the jump. Amd looks like a surer horse to me in the matter, although their crap economical state, i dont expect them to go under in recent time.

If they only supported DX11 then how do you explain them having DX12 results?
And, as I said before, how does a single game in pre-alpha correlate to any sort of final results?

Cheers

Click to expand...

Nvidia: A
Nvidia’s Maxwell GPU is power efficient and fantastically overclockable, with a hard-to-beat option in each pricing tier.

AMD: B
The new Fury X is strong, but the rest of AMD’s aging, rebranded cards are outmatched against the more efficient Nvidia GPUs, though priced well. AMD’s Fury cards coming in the fall may help level the field.

Click to expand...

ozan said:

Specs
ATMINSIDE,

Does anyone really care about power consumption in the high end? Sure it's for the future and the nature and all that stuff but we are talking about a niche market. There is no point in the power efficiency of the card. As long as the cooler is strong enough it doesnt matter.

Yes, people actually do. It makes building in small form factors, but keeping high performance, a reality.

Ofc I mustve said nvidia is completely ignoring dx12, NOT. I think they aren't prioritizing it as high as they should. And I never said there was ever any benchmarks relating to this subject but it was rather abstract, which you have confirmed. There is no real world proof or product yet and all we are doing is expecting something we perceive will happen.

And ED, as I said in a previous post that I am NOT talking about the current generation. Just played rise of the tomb raider. That game wont run very high textures on 4gb cards. The required memory for the cards are increasing at an ever rising rate. Bigger memory will need faster memory and faster bus.

I'm running very high on 5760x1080 on a 4GB GPU, I have no issues with textures popping in/out.
There's a reason we suggest a 980Ti at minimum for 4K and a 980 at minimum for 1440p.

Nv may say they are doing dx12_1. There is a ton of things they said. Like some certain 3.5 gb card that was sold as 4 gb in this gen if my memory serves me right. In the end, it's nvidia that screws over their customers who doesnt upgrade to current gen with new drivers.

You mean the one that still has 4GB of VRAM and, even though the last 0.5GB is slower, is still much faster than backing up to system RAM?
Also, there are absolute slews of people out there that own 970's and are incredibly happy with them. There's nothing in the price range from AMD that competes with it.

Click to expand...