• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
What's happening at BTR *16*
#61
(09-16-2018, 05:47 AM)gstanford Wrote: I like progress too.  The right sort of progress at the right time.

To suit you. Yes, I get it. But I am talking about progress for everyone else.

Gamer4
#62
Vega 64 has more TFLOP than a gtx2080ti , it means nothing in pure gaming performance. Do you really think a gtx2080 ti will only be 19% faster than a gtx1080ti?
#63
I'd suggest you wait for the reviews instead of posting your pre-launch worst case scenarios

[Image: SG.jpg]
#64
(09-16-2018, 01:35 PM)gstanford Wrote: I don't care what the reviews say.

So it's a personal issue. Understood. It's too expensive. You wanted something less forward-looking and more traditional. Well, Nvidia didn't agree with you and they are betting their company on hybrid ray tracing.

Drinks
#65
(09-16-2018, 01:48 PM)gstanford Wrote: This is nvidia trying to force the market down their path instead of listening to what the consumers actually want.  It won't end well for them.

Quote:“If I had asked people what they wanted, they would have said faster horses.”


No doubt they wanted cheaper horses instead of automobiles. Consumers want progress but they don't know have any clue how to achieve it. Much like you being a perfect example ... don't buy Turing. No one cares if you do or don't ...
. . . You are still on Windows 7 on 1080P - RTX ain't for you; it's too expensive and too power-hungry and it won't accelerate your DX9 games very well. Get over it

Crazy_pilot
#66
(09-16-2018, 02:25 PM)gstanford Wrote: Turing isn't progress and whatever the fuck it does isn't RayTracing either.

Another little gem from the Anandtech deep dive was that developers aren't using the AI denoising in their raytracing games (which makes them slower).

Clearly the developers don't buy nvidia's AI bullshit either and most RayTracing games that come out will run on GPU's older than Turing (using DXR not nvidia's proprietary lock in crap).

You are the antithesis of progress. Stuck on Win 7 playing old games at 1080P, paying double for your hardware Down Under, so your posts are self-serving to begin with.

Clearly Turing is doing Ray Tracing in a simplified and hybrid manner using AI to clean up the images, and the result is *still* miles better than rasterization.

We are at the very beginning of ray tracing in real time in games and I applaud Nvidia for making the effort to betting their company on the outcome. I do not know if they will be successful or not. I understand why they did what they did and their move will be seen as a game changer for them - for better or worse. They brought compute to GeForce and Turing uses massive and expensive dies.

Nvidia's proprietary 'crap' is for DLSS - to give basically advanced and free AA to games where devs implement them. And to give a big incentive to the devs, Nvidia offers to optimize and implement DLSS for them. I believe 25 games have already announced they will use DLSS. Ansel is not a failure - neither is ShadowPlay - they are used by a lot of GeForce gamers; clearly Nvidia is leveraging their 80% marketshare. I cannot predict that they will be successful with RTX or not - but I wish them well ... and I do see that they are pushing the photorealistic envelope further and hope it will also impact VR development.
Gamer4
#67
And for all of your whining, negativity, bitterness, hate on Nvidia, and sour grapes; hybrid RTX still looks significantly better than anything that rasterization can accomplish today. It just *looks* better. And you are right, devs are not fools. They know how to enable effective and cheap RTX for a growing market with little effort. Your quote "...at the moment" is meaningless.

Devs and consumers also know that there always has to be a "first generation" for everything that is introduced and it tends to be expensive. Nvidia is taking a major risk to move the state of the gaming art forward while you sit on your ass booing from the sidelines.
Orc

We will see if your dire predictions come to pass or not. I am thinking that Nvidia will succeed with RTX being adopted fairly quickly. Let's revisit this thread later after we see trends and if promises are kept or not and if $1200 is too much for a consumer card. I will only agree to disagree with you on RTX
Paladin
#68
Gstandford,
20% more cuda performance?
Where did you get that figure from?
I calculate at least 35% more .
Gddr6 rescued Nvidia?
25% better memory compression also helps.
Do your homework.
You sir need to wait for reviews before you come to conclusions..
#69
I'm surprised apoppin had the time and patience to respond to all of that. We will see how all this stuff goes. The price is clearly the problem. I am hopeful that the prices will drop by $100-200usd across the board in a few months. Once it does I will be tempted.
#70
(09-16-2018, 11:26 PM)gstanford Wrote:
(09-16-2018, 07:49 PM)happy medium Wrote: Gstandford,
20% more cuda performance?
Where did you get that figure from?
I calculate at least 35% more .
Gddr6 rescued Nvidia?
25% better memory compression also helps.
Do your homework.
You sir need to wait for reviews before you come to conclusions..

The CUDA performance figures come from the Anandtech Deep Dive article.  I gave direct quotes all 3 times (and 20% is optimistic given 19% was the actual maximum improvement) Are you suggesting they are liars?!


Forum Jump:


Users browsing this thread: 5 Guest(s)