• 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.
They are optimised for both, its simply that nVidia are lacking some A-Sync features.

nVidia have had 6 months to get all they can out of Doom Vulkan, it is what it is....

Your Right it is what it is, but what is it? Imo it is pointless for Nvidia to optimise for vulkan when opengl performs similar on Nvidia cards.

Amd of course is well worth optimising vulkan, vulkan needs to be careful it does not fall into the realms of an amd only api or else it will be dead in the water only catering for 30% of gamers.
 
What a legend, Now I'm happy :D:D:D:D:D:D :D

How did you do the cranked big grin? i love it :D:D:D:D:D:D

Your Right it is what it is, but what is it? Imo it is pointless for Nvidia to optimise for vulkan when opengl performs similar on Nvidia cards.

Amd of course is well worth optimising vulkan, vulkan needs to be careful it does not fall into the realms of an amd only api or else it will be dead in the water only catering for 30% of gamers.

Vulkan is a gain for all, what you never see from any mainstream reviewer, just as with Mantle before it and to some extent DX12 is its ability to elevate low end or old CPU's to be much much much better than they are in DX11 and OpenGL.

You run a GTX 1060 with a Phenom II or Core Duo the performance difference between OpenGL and Vulkan will be very pronounced.

Besides that even for nVidia Vulkan has higher performance than OpenGL even on the fastest CPU's.

The reason for AMD better than nVidia performance in Vulkan is simply that AMD have better Asynchronous shading.
 
The Fury X was good, very good in fact, but that stupid 4GB was a limiting factor. Had they put 8GB on, it would have been the perfect 4K card.

If the Fury had had 8gb's or ram and 40-50 % more performance then it would have been the perfect 4k card, actually with 8gb's of memory and 40-50% more performance it would have been Vega! Drums please....

How did you do the cranked big grin? i love it :D:D:D:D:D:D

It's a size thing :D
 
I compared my Titan X Maxwell to my Fury X at 1440P and it was a bit one sided at that res and when overclocking the 2, the gap was even more but had the FX had 8GB, it would have been a fantastic card for 4K resolutions and giving the Maxwell Titan some what for in many games.

Trouble is, the Fury X isn't powerful enough on it's own to power a lot of the latest games in 4k (unless you want to turn all the settings down but who wants to do that!), so its HBM was uselss; Great for 4k, but not enough of it for 4k...on a GPU that wasn't powerful enough for 4k.

Again, like always for AMD recently, it is too much, too soon (rather than too little, too late :D).

Of course the real reason for using HBM was to get the power draw down to just about acceptable levels, but it was expensive and 4gb wasn't enough so it held back the card somewhat as well. AMD were between a rock and hard place with Fiji to be honest.
 
The problem is now that prices have been elevated like they are they're unlikely to go down much, AMD have watched Nvidia push prices up with the 10 series and they will want a piece of the pie so they'll follow suit, They may undercut Nvidia but not by much..

If AMD do this then they are falling prey to what usually happens when they have had competitive cards at slightly better prices. More nVidia cards get sold because they have the better brand image. I think AMD know what they need to do because they are listening now. It's on them to do well.
 
If AMD do this then they are falling prey to what usually happens when they have had competitive cards at slightly better prices. More nVidia cards get sold because they have the better brand image. I think AMD know what they need to do because they are listening now. It's on them to do well.

short memory wont help.

as we seen Ryzen allows a 40-55% ipc gain when everyone said amd was doomed. same thing will happen with gpu and graphics.
then obviously I will remember that post u made. :D
 
They are optimised for both, its simply that nVidia are lacking some A-Sync features.

From what I've been told there is improvements that could be made on the nVidia side it just hasn't been a priority - one of the complications being that you have to do something different for Kepler versus Maxwell and Pascal and even Pascal has some potential room for improvements Maxwell doesn't so the current implementation is basically the one that has the least negative penalties when used indiscriminately.
 
From what I've been told there is improvements that could be made on the nVidia side it just hasn't been a priority - one of the complications being that you have to do something different for Kepler versus Maxwell and Pascal and even Pascal has some potential room for improvements Maxwell doesn't so the current implementation is basically the one that has the least negative penalties when used indiscriminately.

Right, another reason i want out of the Nvidia eco system, if they can't be bothered to optimise games properly then i can't be bothered to give them my money.

Typical useless........ i do miss AMD for user support.
 
short memory wont help.

as we seen Ryzen allows a 40-55% ipc gain when everyone said amd was doomed. same thing will happen with gpu and graphics.
then obviously I will remember that post u made. :D

Flopper i do love your posts ;) I wish i could be as optimistic as you but it hurts too much when my optimism is killed. BUUUUT i do really hope AMD come out really well with both CPU and GPU departments! I need a upgrade path man! :D
 
Flopper i do love your posts ;) I wish i could be as optimistic as you but it hurts too much when my optimism is killed. BUUUUT i do really hope AMD come out really well with both CPU and GPU departments! I need a upgrade path man! :D

Much easier to be on the side of caution so if it turns out good it's a pleasant surprise. With Fury X and RX480 i was to optimistic and felt a little let down. More so with Fury X.
 
Right, another reason i want out of the Nvidia eco system, if they can't be bothered to optimise games properly then i can't be bothered to give them my money.

Typical useless........ i do miss AMD for user support.

Agree just another one to tick on the box

I have been wanting to jump ship for a while now even though i am happy with my nvidia card re performance.. but what i am not happy about is nvidia as a company ....
over the past few years i have become to dislike nvidia as a company

i am surprised they have the mind share they have

So if vega can deliver then they have my money along with getting a Ryzen cpu
 
Right, another reason i want out of the Nvidia eco system, if they can't be bothered to optimise games properly then i can't be bothered to give them my money.

Typical useless........ i do miss AMD for user support.

Hang on, isn't the optimising for the different Nvidia ranges due to changes in tech and doesn't that come down to the developer optimising for each one if they so choose as Nvidia improve and change their product? You want advancements in tech but don't want it to affect how it is done so stagnate the way cards are produced to make an eco-system work as long as you deem it is required too?

The difference here is actually AMD have just been stagnant (in some ways) working on the same system and improving that. Nvidia have not to the same extent and tried different things to get more from their cards.

I am not worried about which brand I use, just trying to suggest that the support for Doom comes from the developers if they want to push Nvidia but have stated they are not because Nvidia already runs that well and AMD needed the push to improve it to match what Nvidia was already doing?

Surely overall that is a win for Nvidia as they didn't need to go down a whole you API path like they did for AMD?
 
Last edited:
Status
Not open for further replies.
Back
Top Bottom