• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD VEGA confirmed for 2017 H1

Status
Not open for further replies.
silversurfer said:
8k for when you gonna game via a projector onto the side of a building :D
Click to expand...

I struggle to see the point in even mentioning 8k, one minute it's VR and the next it's this.

The masses are on 1080p and those with higher resolutions know the latest and greatest hardware only just manages 4k so 8k's a definite no go unless you want to run Pac man, CS go at 60 or Halflife..
I'm getting tired of the constant push for us consumers to look forward when they're struggling to even put anything on the high end market.

TNA said:
Erm, jumping the gun a bit there are we not AMD? Let's get a proper 4K card out first before we start talking about 8K? lol :p

Love the sound of Freesync 2 though :)
Click to expand...

The good thing about Freesync 2 is they control who uses the name and they also insist on any monitors using it having decent support.
 
Stole this from AT forums

I will repost this from Sebbbi, so that everyone can read it:
https://forum.beyond3d.com/posts/1973875/

Just wanted to clarify that I meant AMD GCN2 (consoles) vs Nvidia's latest (Maxwell/Pascal). AMD PC GPUs have also improved since GCN2.

Improvements for general performance:
- GCN3 introduced delta color compression. Including ability to sample/load compressed textures without decompress step.
- GCN3 improved geometry tessellation performance
- GCN4 improved geometry performance in general (including fast strips, primitive discard, etc).
- GCN4 improved delta color compression.
- GCN4 added instruction prefetch (reduces pipeline latency, again helps with geom bottleneck).
- GCN4 improved async compute scheduling (GPU side)

GCN5 (Vega) adds these general performance improvements:
- L2 cache includes L2 ROP cache (L1 ROP caches under L2). Don't need to flush caches between pixel shader passes.
- Tiled rasterizer. Reduces overdraw, bandwidth and makes ROPs more efficient in general.
- Improved geometry pipeline (including proper load balancing, up to 2x higher peak throughput)
- General purpose memory paging system

(I didn't list features that don't bring performance improvements without programmer intervention)

All of these improvements mean that GCN5 should run general purpose pixel/vertex shader code much better than GCN2. GCN5 has most of the same tricks that are seen in modern Nvidia GPUs. There are nice compute improvements as well, but they need special programmer support (DPP, SDWA, FP16). We will see the real impact of these improvements when DX12 SM 6.0 becomes available. Doom is already using these features with Vulkan, resulting in nice gains.
 
opethdisciple said:
If I move over from Nvidia I am doing it for one reason and one reason alone. Better DX12 and Vulkan support. I feel let down by Nvidia in these areas.

Yes I have not used a 1000 series card and am basing these assumptions on my 970 but I have observed that the 1000 series have also not been great DX12 cards and with Ryzen these assumptions seem to have rising to public opinion also.

So my question to the Red team is, in the games that DX12 is an option have you been able to use it, with little to no issues?

In all DX12 enabled games with my 970, performance has regressed. In Doom with Vulkan it wasn't as good as OpenGL.
Click to expand...

It's not a problem to compare with a 970. It's been a slight running joke with the 1000 series, but it really is basically heavily overclocked Maxwell.

Pascal has all the strengths and weaknesses of the 900 series, and I've never seen an example showing "look here, this thing here Pascal is great at and Maxwell sucks at".
 
SiDeards73 said:
All of these improvements mean that GCN5 should run general purpose pixel/vertex shader code much better than GCN2. GCN5 has most of the same tricks that are seen in modern Nvidia GPUs. There are nice compute improvements as well, but they need special programmer support (DPP, SDWA, FP16). We will see the real impact of these improvements when DX12 SM 6.0 becomes available. Doom is already using these features with Vulkan, resulting in nice gains.
Click to expand...

So another fine wine tech then :p

As long as performance is there from day one and price for performance is good, I am happy. At least we know as a bonus performance will get better with time :D
 
opethdisciple said:
The 1080ti is nothing 'new'. It's just more of the same just faster.
Click to expand...
At the end of the day, Vega wont necessarily change the game either and will just be new, faster GPU's. Most people aren't generally super interested in tiled rasterizers, they just want to know what kind of improvement the GPU as a whole will give in performance.

I think the excitement over Vega is the competition aspect. People want AMD to come out with something that brings competition to Nvidia at the higher end. It has implications for everybody, regardless of if you have a preference of brand or not.
 
Seanspeed said:
At the end of the day, Vega wont necessarily change the game either and will just be new, faster GPU's. Most people aren't generally super interested in tiled rasterizers, they just want to know what kind of improvement the GPU as a whole will give in performance.

I think the excitement over Vega is the competition aspect. People want AMD to come out with something that brings competition to Nvidia at the higher end. It has implications for everybody, regardless of if you have a preference of brand or not.
Click to expand...

Vega will definitely be technically better, but faster?? The aftermarket Ti's are fast as all hell, I'm doubtful!
 
Status
Not open for further replies.
Back
Top Bottom