• 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.

NVIDIA ‘Ampere’ 8nm Graphics Cards

TNA said:
Unless paid for, why will they spend extra resources to code for Nvidia’s custom implementation? Nvidia will have to pay I would imagine. Nvidia will need to provide something MUCH better than the RT that the 2000 series provides in order to stand out and have devs want to code for that without being paid extra to do so.
Click to expand...



Because nvidia doesn't have a custom implementation. Nvidia supports the open standard in Vulcan and the MS standard in DX.

Your question is the wrong way round. Why would a dev support a non-standard customized API form AMD when they have a far smaller user base and would have spent the last few years using the Microsoft standard?
 
D.P. said:
Navi 10 is built on 7nm DUV, Ampere will be built on 7nm EUV, that is basically the same jump as going from 16nm to 7nm DUV.
Moreover, 7nm EUV offers higher yields so bigger chips are more viable.
Click to expand...

Yup, if EUV 7nm+ gives 20% higher transistor density, the figure would increase to ~49Mtr/mm2.
 
D.P. said:
Because nvidia doesn't have a custom implementation. Nvidia supports the open standard in Vulcan and the MS standard in DX.

Your question is the wrong way round. Why would a dev support a non-standard customized API form AMD when they have a far smaller user base and would have spent the last few years using the Microsoft standard?
Click to expand...
Seems I got it jumbled up. What I am trying to say is, they will build games for consoles which will be for AMDs implementation. Will they then go and change it just for PC market unless paid extra by nvidia to do so?
 
TNA said:
Seems I got it jumbled up. What I am trying to say is, they will build games for consoles which will be for AMDs implementation. Will they then go and change it just for PC market unless paid extra by nvidia to do so?
Click to expand...

surely Microsoft will use their own implementation in the new console? No idea about Sony.
 
Stanners said:
surely Microsoft will use their own implementation in the new console? No idea about Sony.
Click to expand...
We will see how it goes. Just hoping that the 3000 series offers a lot more in terms of RT as the 2000 series as we all can see has been very lacking. Would be nice to be able to have it on while playing Cyberpunk 2077 with something like say a 3070 without needing to spend silly moneys on a 3080Ti.
 
Mauller said:
7nm+ is only EUV for 4 BEOL layers, it's not a full EUV node. 5nm is the full EUV node from TSMC.
Click to expand...
Regardless, 7nm EUV has 10% less power and 20% greater density than 7nm DUV, all while be cheaper and with higher yields.

5nm EUV is actually a smaller tweak of 7nm EUV
 
TNA said:
We will see how it goes. Just hoping that the 3000 series offers a lot more in terms of RT as the 2000 series as we all can see has been very lacking. Would be nice to be able to have it on while playing Cyberpunk 2077 with something like say a 3070 without needing to spend silly moneys on a 3080Ti.
Click to expand...


MS will support a verison of DX and the RTX AP they developed.

I think you jumped to the conclusion that Nvidia must have created a proprietary API when they have done the exact opposite by working with Microsoft and the Vulkan community to create standard APIs.

I would be very surprised if AMD did anything other than support MS DX RTX and Vulkan RTX.


As to RTX performance, I expect the 3000 series will at least double performance but triple performance should be possible. Nvidia allocated very few transistors to RTX on Turing because they were limited on the 12nm process. Most of the increase in die area comes from Turing's internal changes for high scaling and efficiency. That will directly translate to higher performance of Ampere when the CUDA cores a cranked up.Turing is actually a very future looking change, but was limited to the fab process in realizing the full scope of improvements. This will ultimately mean there is an additional scope for more RTX cores. Only 5% of the Turing die is dedicated to RTX, this could increase to 10%. Then this 10% has about an 80% higher density, running around 50% faster clocks. And then there can be lots of architectural changes to improve efficiency. RTX in Turing was first generation, so another 30% or more performance could be seen via generational changes.
 
beany_bot said:
I've played games with decent ray tracing (my best mate has a 2080Ti). It's nice, for sure. looks ok. But I really just see it more as a graphics slider that looks "decent" but TANKS performance. Plenty games I have thought... oh wow is that ray tracing!? and its not. the difference just isnt nearly as amazing as people are getting swept up in it belive. It just a graphics option like hairworks or ambient occlusion. unfortunately its a graphics option that literally ruins gameplay FPS.
Click to expand...
What games have you played? I really like it in Metro and really does add to the immersion. Frames take a hit but with some graphic tweaking, my 2080Ti copes perfectly well at 3440x1440 and no stutter or slow downs that I ever notice.
 
beany_bot said:
There are tons of figures out there. Turing has been a flop compared to maxwell, pascal, etc etc.
Still profitable yes but not nearly as much as previous gen. Sales are way less than Pascal.
Click to expand...

Turing has been a success for NVidia

For
Faster conventional performance than anything before it.
RTX and DLSS features (yes the fps take a hit but the opposition has no built in hardware at all).
Improved cooler.


Against
Price.


Turing sales can not be compared to Maxwell and Pascal as there are fewer SKUs for the newer architecture and they have not been available for as long as the two older ones.

I am not happy about the price of Turing at all but having said that there is nothing better than the RTX Titans for gaming. I have been doing a lot of gaming lately so I think it was worth it to me to get the Titans, I know people who will spend more on cars or in the pub but that is their pastime.
 
D.P. said:
MS will support a verison of DX and the RTX AP they developed.

I think you jumped to the conclusion that Nvidia must have created a proprietary API when they have done the exact opposite by working with Microsoft and the Vulkan community to create standard APIs.
Click to expand...

Is called DXR from DirectX Ray Tracing. Is not RTX which is Nvidia proprietary ray tracing API.

You need to understand there are 2 APIs atm. DXR (Microsoft DX12.1 compatible cards can use it) & RTX (Nvidia proprietary supported by Volta & Turing)
Don't confuse the both.

RTX API is been used for the support of Wolfenstein Vulkan engine.
DXR is been used on BF5 & Metro Exodus.

Update. Forgot Intel's hybrid Ray Tracing going to be used on World of Tanks with the next patch
 
Last edited:
Kaap what do you do with your old high end cards do you collect them?

I was thinking back to the days of weird cards like the Asus MARS that was about a grand. Now the 2080ti STARTS at a grand lol
 
Kaapstad said:
I am not happy about the price of Turing at all but having said that there is nothing better than the RTX Titans for gaming. I have been doing a lot of gaming lately so I think it was worth it to me to get the Titans, I know people who will spend more on cars or in the pub but that is their pastime.
Click to expand...

Are you putting them to work properly? What have you been playing Kaap?
 
You must log in or register to reply here.
Back
Top Bottom