AMD NAVI Radeon 5700 Product Announcement thread

It's key if you want to be able to turn it into a mobile gpu for laptops. Then efficiency is king. With the 7nm process advantage I think it's fair to say we all hoped for better.

 

Sadly radeon group is going under while their cpu group is rising. We are waiting and waiting and looks like there is no another 7950/7970 on sight. Nvidia will have like ten rtx cards (regular, super) while amd have two navi cards, overpriced on top of it.

 

I dont know about that. I think AMD is in a better position to undercut Nvidia on the price front, esp with half the die sizes vs RTX cards. Although I dont think we will see repeat RX 570/580 pricing, they are in a much better position to gain the bang for the buck crown just like with Polaris.

 

Interesting time ahead no doubt. I think Navi is arch to break it or make it for AMD. There will be probably higher number of cu's ver coming oit later this year. Intel is also joining the fight next year.

2020 gaming systems :

AMD = PS5, Xbox scarlet, Stadia, Samsung phones, Switch 2 (rumor), PC

Nvidia = PC

 

That's so the bosses can hold it better when showing it off, compare page 2's picture of AMD's boss

 

Hmm, Polaris replacement for 379-449€ -> Deal of the day
Should be 299-349€ BTW
ATI now selling Vega 64 & 2070

3xFan
-> https://www.mindfactory.de/product_...Aktiv-PCIe-3-0-x16--Full-Retail-_1235216.html
or
3xFan
-> https://www.mindfactory.de/product_...xtreme-Core-Aktiv-PCIe-3-0-x16--_1281215.html

 

To be fair this card is much faster than any Polaris. In DX12 it should be nipping the heals of the 1080ti/2080. For the record I feel $399/$329 should be the prices.

 

Well you know how this goes. When they see nvidia is faster again they'll reduce the prices like always.

 

AMD could probably give 2080ti a run for its money with 64CU Navi .Power is the only concern when they get there, as die size would be well under control even at 64CU.

 

I think both need to be £50 cheaper.
Be nice is they released one with 128 compute units with 32 gig of hbm2
they could but choose not to.

 

Man, that would be 30B transistors. Over 50% more than 2080Ti. And 750mm^2 die size. Cost on 7nm would not be friendly same way as yields.

 

They kinda did with radeon pro duo. Thou in fairness that's two 64CU Vegas "glued together". Sorry I had to.

 

Only if the cards are compared on different nodes ... competition would be skewed again when Nvidia is at 7nm.

 

I don’t give a rats ass what node each chip is on all I care about is performance.

 

Prior release of Turing, nVidia had better performance per transistor per clock.
With Turing, they degraded themselves due to investment which did not exactly improve classical rendering methods. And as result Vega is practically equal with Turing in this comparison.
But Navi in its 1st iteration is better in this department. And we can only expect better performance per transistor per clock.

Issue at hand is that AMD is AMD and they like to use available clocks and therefore set higher TDP/TBP. On other hand Navi can do 2GHz and clocking it out of the box to just 1500~1600 MHz must look like quite some waste.

So at time nVidia delivers 7nm EUV GPUs next year, AMD will deliver their alternative. And here is what I expect:
nVidia will double down on RTX capability which will again weaken their performance per transistor per clock.
AMD will deliver final Navi configuration which will improve again performance per transistor per clock.

AMD will have quite some advantage in terms of manufacturing cost of GPUs vs. delivered performance. It may come again at cost of higher power consumption. But that remain to be seen. And we do not really know real power draw characteristics of 5700 XT and what we can do about it.

 

thats not how any of this works.

 

That's exactly how it works.

 

I think what Astyanax wants to say is that he gives a f**** about the guts of the card itself as long as it delivers on review charts. You are good with half card dead/sleep while gaming (rtx cores) if the other halfs performance is better than the competition "full card".

Thruth is you pay for that "dead" half and the company that build and desing it too, so you are wasting money on transistors that deliver 0 performance gain.

 

hey, guess what.

your gpu always has inactive components at some point or another.

not all games use tesselators
not all games use hull or geometry shaders.
RT cores being idle is nothing new.

neither is architecture ignorance by consumers.

 

Everything has inactive components at some point in time, even human brain, thats not what we are discussing here.

A simple fact about rtx components atm is that many of the low rtx cards bought by consumers will die with the rtx cores as new because they were never used. Maybe not in the 2080/2080ti space, but surely in some 2070 and 90% of the 2060.