Monday, May 27th 2019

AMD Announces Radeon RX 5700 Based on Navi: RDNA, 7nm, PCIe Gen4, GDDR6

AMD at its 2019 Computex keynote today unveiled the Radeon RX 5000 family of graphics cards that leverage its new Navi graphics architecture and 7 nm silicon fabrication process. Navi isn't just an incremental upgrade over Vega with a handful new technologies, but the biggest overhaul to AMD's GPU SIMD design since Graphics CoreNext, circa 2011. Called RDNA or Radeon DNA, the new compute unit by AMD is a clean-slate SIMD design with a 1.25X IPC uplift over Vega, an overhauled on-chip cache hierarchy, and a more streamlined graphics pipeline.

In addition, the architecture is designed to increase performance-per-Watt by 50 percent over Vega. The first part to leverage Navi is the Radeon RX 5700. AMD ran a side-by-side demo of the RX 5700 versus the GeForce RTX 2070 at Strange Brigade, where NVIDIA's $500 card was beaten. "Strange Brigade" is one game where AMD fares generally well as it is heavily optimized for asynchonous compute. Navi also ticks two big technology check-boxes, PCI-Express gen 4.0, and GDDR6 memory. AMD has planned a July availability for the RX 5700, and did not disclose pricing.
Add your own comment

202 Comments on AMD Announces Radeon RX 5700 Based on Navi: RDNA, 7nm, PCIe Gen4, GDDR6

#101
GoldenX
It might come in a future update.
Posted on Reply
#102
FordGT90Concept
"I go fast!1!11!1!"
Minus Infinity said:

Then why do the LG 2019 C9 TV's have HDMI 2.1. How did they manage to get that done for an already released product? They even announced 2.1 support 6 months ago.
New GPU architectures take about 5 years to go from design to commercial product. HDMI 2.1 has a lot of significant changes in it that were too much to accommodate too late in development. The most important features (like VRR) AMD already does over HDMI 2 to displays that support it.

The decoder chips in TVs are vastly more simple than GPUs.


Edit: If this is the TV you're talking about it looks like the only HDMI 2.1 feature they added was eARC...which is simple.
Posted on Reply
#103
StudMuffin
LocutusH said:

I dont see HDMI 2.1? Why?
I don't think it really matters as long as the HDTV is a half-way decent HDTV with HDR,etc..most of these 2018/2019 4K HDTV's that are mid range and up do HDR and true 120hz, what people want and like is 4K resolution at 120Hz, and maintains your 4:4:4 subsampling and is still plenty smooth.... If that is the case..You'll be able to get all those important benefits of HDMI 2.1 using the current HDMI 2.0 . You can get these important "gaming features" through current HDMI 2.0, again, as long as the 4K HDTV is decent. The only reason a gamer should care about having a GPU with HDMI 2.1 is if they are wanting to run their games at 8K at 60fpz/120fps and thats a pipe dream right now lol.

So if a gamer wants to have their PC connected to their 4K HDTV and run their PC games at 4K rez at a maximum of 120Hz, and maintains your 4:4:4 subsampling and is still silky smooth, this can be done with HDMI 2.0, as long as the 4k HDTV is decent, that has a 120hz true processor inside and can do that while keeping the 4 4 4 subsamp, HDR,etc,etc, Some of these 4k Sets claim it can do 120hz but honestly its not true 120hz, its 60hz but most mid-range 4k sets, that have come out in the last few years or for sure this year..can do this.
Posted on Reply
#104
TechLord
Navi 5700 will be 10%-15% faster than Rtx 2070

Navi 5800 will be 10%-15% faster than Rtx 2080

Navi 5900 will be 10%-15% faster than Rtx 2080ti and the 5900 series will have (7nm+ Arcturus hardware accelerated Ray Tracing) features, and that specific Gpu will launch mid 2020


Navi 5700 will be their entry level Navi mid range Gpu which will beat a 2070 launching this July, and then in 2020 we'll get the 5800 & 5900 Gpu's which will beat the 2080 & 2080ti respectively


The 5700 & 5800 based Pc parts will be standard off the shelf Gpu's without any specific hardware accelerated Ray Tracing features baked in, however the next Gen consoles will get fully custom Navi based Gpu's with the 5900 hardware accelerated 7nm+ Arcturus Ray Tracing features built into the silicon as these next Gen console's will be the real deal Holyfield people.


Xbox Anaconda will use the full custom 5900 Gpu with advanced Ray Tracing Arcturus features....


Ps5 will use a custom 5800 Gpu with advanced Ray Tracing Arcturus Features....


Xbox Lockhart will use a custom 5700 Gpu with advanced Ray Tracing Arcturus Features....


These consoles will perform far higher than the +10% figures of the Pc parts above due to the way consoles work with deep console level optimization, and closed loop ultra low level to the Metal console Api's......


These console versions of Navi will be custom parts which are far more advanced than the Pc parts........


Next Gen console's Xbox Anaconda & Ps5 will absolutely murder the performance of an Rtx 2070 level Pc as Anaconda will be faster than an Rtx 2080ti, and ps5 will be faster than an Rtx 2080, and Xbox Lockhart will be faster than an Rtx 2070
Posted on Reply
#105
londiste
Vayra86 said:
I'm no fan either of large GPU die percentages dedicated to just RT performance
That percentage is surprisingly small.
Vayra86 said:
The real question is what the market will accept in terms of visual gain versus additional cost / performance hit. And that is an answer nobody has, but Turing so far isn't going like hotcakes, which is a sign. In that sense, if we can see in-game, live footage of that Crytek implementation adding to visual quality at minimal performance cost, that is the real game changer.
As I noted above, what Vega56 does in Neon Noir, GTX1080 can do in Battlefield V. This is the level of performance expected today from RT running on shaders.
Vayra86 said:
I think the more interesting development with hardware solutions for RT is how well it can be utilized for other tasks. That will make RT adoption easier. Nvidia tried something with DLSS, but that takes too much effort.
DLSS is not RT at all. Nvidia tried denoising on Tensor cores which is similar but different from DLSS and they have been suspiciously quiet on whether actual games are using it this far.
medi01 said:
DXR works with different structures, Crytek is voxel based, DXR is not.
So there goes the "could be used" aspect of it, because, wait for it, "specialized hardware" is not known for being flexible.
BHV traversal can be run on RT cores for voxels. Intersection would probably need meshing of the voxels.
medi01 said:
We can have all those visuals today, with helluva shader work, the main point of RT gimmick (and it's nothing beyond it, for F sake, most of RT-ing is denoising at this point) is to achieve reflections/illumination/shade with smaller effort.
I think you underestimate how much smaller this effort is.
medi01 said:
For game developers to do it, one simply needs to have large enough "RT user base". And this is why Crytek's take on the problem is so much better than NVDA's.
You do realize that "Nvidia's approach" will give a considerable speed boost to "Crytek's take"? The truth of the matter is that there are no different approaches as such. This is all ray tracing.
StudMuffin said:
I don't think it really matters as long as the HDTV is a half-way decent HDTV with HDR,etc..most of these 2018/2019 4K HDTV's that are mid range and up do HDR and true 120hz, what people want and like is 4K resolution at 120Hz, and maintains your 4:4:4 subsampling and is still plenty smooth.... If that is the case..You'll be able to get all those important benefits of HDMI 2.1 using the current HDMI 2.0 . You can get these important "gaming features" through current HDMI 2.0, again, as long as the 4K HDTV is decent. The only reason a gamer should care about having a GPU with HDMI 2.1 is if they are wanting to run their games at 8K at 60fpz/120fps and thats a pipe dream right now lol.
HDMI 2.0 cannot do 4k@120Hz, much less at 4:4:4 and VRR.
Posted on Reply
#107
FordGT90Concept
"I go fast!1!11!1!"
It's a mid-range chip. It was never intended to be top dog. Radeon VII remains AMD's top product until Arcturus debuts.

I suspect Navi only has 8-10 billion transistors compared to 18.5 billion in RTX 2080 Ti and 13.5 billion in Radeon VII. The fact it is knocking on Radeon VII's door with GDDR6 and much fewer transistors is a testament to RDNA's design.
Posted on Reply
#108
Valantar
TechLord said:

Navi 5700 will be 10%-15% faster than Rtx 2070

Navi 5800 will be 10%-15% faster than Rtx 2080

Navi 5900 will be 10%-15% faster than Rtx 2080ti and the 5900 series will have (7nm+ Arcturus hardware accelerated Ray Tracing) features, and that specific Gpu will launch mid 2020


Navi 5700 will be their entry level Navi mid range Gpu which will beat a 2070 launching this July, and then in 2020 we'll get the 5800 & 5900 Gpu's which will beat the 2080 & 2080ti respectively


The 5700 & 5800 based Pc parts will be standard off the shelf Gpu's without any specific hardware accelerated Ray Tracing features baked in, however the next Gen consoles will get fully custom Navi based Gpu's with the 5900 hardware accelerated 7nm+ Arcturus Ray Tracing features built into the silicon as these next Gen console's will be the real deal Holyfield people.


Xbox Anaconda will use the full custom 5900 Gpu with advanced Ray Tracing Arcturus features....


Ps5 will use a custom 5800 Gpu with advanced Ray Tracing Arcturus Features....


Xbox Lockhart will use a custom 5700 Gpu with advanced Ray Tracing Arcturus Features....


These consoles will perform far higher than the +10% figures of the Pc parts above due to the way consoles work with deep console level optimization, and closed loop ultra low level to the Metal console Api's......


These console versions of Navi will be custom parts which are far more advanced than the Pc parts........


Next Gen console's Xbox Anaconda & Ps5 will absolutely murder the performance of an Rtx 2070 level Pc as Anaconda will be faster than an Rtx 2080ti, and ps5 will be faster than an Rtx 2080, and Xbox Lockhart will be faster than an Rtx 2070
Gotta love it when brand-new accounts come in and spew heaps of incredibly optimistic and entirely unsourced speculation. The most trustworthy information there is.

I'm hopeful for Navi, but this is just silly.
Posted on Reply
#109
Vayra86
TechLord said:

Navi 5700 will be 10%-15% faster than Rtx 2070

Navi 5800 will be 10%-15% faster than Rtx 2080

Navi 5900 will be 10%-15% faster than Rtx 2080ti and the 5900 series will have (7nm+ Arcturus hardware accelerated Ray Tracing) features, and that specific Gpu will launch mid 2020


Navi 5700 will be their entry level Navi mid range Gpu which will beat a 2070 launching this July, and then in 2020 we'll get the 5800 & 5900 Gpu's which will beat the 2080 & 2080ti respectively


The 5700 & 5800 based Pc parts will be standard off the shelf Gpu's without any specific hardware accelerated Ray Tracing features baked in, however the next Gen consoles will get fully custom Navi based Gpu's with the 5900 hardware accelerated 7nm+ Arcturus Ray Tracing features built into the silicon as these next Gen console's will be the real deal Holyfield people.


Xbox Anaconda will use the full custom 5900 Gpu with advanced Ray Tracing Arcturus features....


Ps5 will use a custom 5800 Gpu with advanced Ray Tracing Arcturus Features....


Xbox Lockhart will use a custom 5700 Gpu with advanced Ray Tracing Arcturus Features....


These consoles will perform far higher than the +10% figures of the Pc parts above due to the way consoles work with deep console level optimization, and closed loop ultra low level to the Metal console Api's......


These console versions of Navi will be custom parts which are far more advanced than the Pc parts........


Next Gen console's Xbox Anaconda & Ps5 will absolutely murder the performance of an Rtx 2070 level Pc as Anaconda will be faster than an Rtx 2080ti, and ps5 will be faster than an Rtx 2080, and Xbox Lockhart will be faster than an Rtx 2070
The Lord has spoken! :roll:
Posted on Reply
#110
Ibotibo01
TechLord said:

Navi 5700 will be 10%-15% faster than Rtx 2070

Navi 5800 will be 10%-15% faster than Rtx 2080

Navi 5900 will be 10%-15% faster than Rtx 2080ti and the 5900 series will have (7nm+ Arcturus hardware accelerated Ray Tracing) features, and that specific Gpu will launch mid 2020


Navi 5700 will be their entry level Navi mid range Gpu which will beat a 2070 launching this July, and then in 2020 we'll get the 5800 & 5900 Gpu's which will beat the 2080 & 2080ti respectively


The 5700 & 5800 based Pc parts will be standard off the shelf Gpu's without any specific hardware accelerated Ray Tracing features baked in, however the next Gen consoles will get fully custom Navi based Gpu's with the 5900 hardware accelerated 7nm+ Arcturus Ray Tracing features built into the silicon as these next Gen console's will be the real deal Holyfield people.


Xbox Anaconda will use the full custom 5900 Gpu with advanced Ray Tracing Arcturus features....


Ps5 will use a custom 5800 Gpu with advanced Ray Tracing Arcturus Features....


Xbox Lockhart will use a custom 5700 Gpu with advanced Ray Tracing Arcturus Features....


These consoles will perform far higher than the +10% figures of the Pc parts above due to the way consoles work with deep console level optimization, and closed loop ultra low level to the Metal console Api's......


These console versions of Navi will be custom parts which are far more advanced than the Pc parts........


Next Gen console's Xbox Anaconda & Ps5 will absolutely murder the performance of an Rtx 2070 level Pc as Anaconda will be faster than an Rtx 2080ti, and ps5 will be faster than an Rtx 2080, and Xbox Lockhart will be faster than an Rtx 2070
No, it won't be happening. Sapphire leaks Navi's price (5700 and 5800). AMD uses Radeon VII which is top GPU. Navi 5700 is %10 faster than RTX 2070 in Strange Brigade. Already, Radeon VII is %20-25 faster than RTX 2080 [B]in Strange Brigade but RTX 2080 is faster than Radeon VII in other games like Witcher 3, AC Odyssey and Rainbow Six Siege. [/B]
My Expections
RX 5700= RTX 2060+%7
RX 5800= RTX 2070
RX 5900= RTX 2070+%10
RX 5900 XT= RTX 2080
Posted on Reply
#111
Valantar
Vayra86 said:

The Lord has spoken! :roll:
The Lord hath spoken.

Can't have inaccuracies on the forums, jeez.
Posted on Reply
#112
medi01
londiste said:

You do realize that "Nvidia's approach" will give a considerable speed boost to "Crytek's take"?
I need to see the receipts.

Vayra86 said:

I suppose you know better than what sources have shown thus far.
I suppose you have sources that others don't, don't hesitate to show them.

PS
And why always so outraged?
Posted on Reply
#113
Frick
Fishfaced Nincompoop
Valantar said:

The Lord hath spoken.

Can't have inaccuracies on the forums, jeez.
This really depends on if you're quoting the Bible or Händel.

Anyway. Is this brand new or not? I see many conflicting arguments. Me I'm cautiously optimistic, in this context defined as "might perform almost as good as they say in a best case scenario".
Posted on Reply
#114
Vayra86
medi01 said:

I suppose you have sources that others don't, don't hesitate to show them.
Sure thing.

https://www.cryengine.com/news/how-we-made-neon-noir-ray-traced-reflections-in-cryengine-and-more#

"However, RTX will allow the effects to run at a higher resolution. At the moment on GTX 1080, we usually compute reflections and refractions at half-screen resolution. RTX will probably allow full-screen 4k resolution. It will also help us to have more dynamic elements in the scene, whereas currently, we have some limitations. Broadly speaking, RTX will not allow new features in CRYENGINE, but it will enable better performance and more details. "

As for 'being outraged'... I think that's more up to your misguided interpretation of the world around you than anything else. The fact you're not aware of the above source speaks volumes. There is no outrage here.
Posted on Reply
#115
Manoa
when crysis doing refleractions in half resolution, you know the game is over
Posted on Reply
#116
medi01
Vayra86 said:

"However, RTX will allow the effects to run at a higher resolution. At the moment on GTX 1080, we usually compute reflections and refractions at half-screen resolution. RTX will probably allow full-screen 4k resolution. It will also help us to have more dynamic elements in the scene, whereas currently, we have some limitations. Broadly speaking, RTX will not allow new features in CRYENGINE, but it will enable better performance and more details. "
Nice (emphasis mine).


As for 'being mad'... I think that's more up to your misguided interpretation of the world around you than anything else. There is no "being mad" here.

Manoa said:

when crysis doing refleractions in half resolution, you know the game is over
Because RTX gimmick is... cough.


RTX shadows are actually way crappier than what is shown above. That's why heavy denoising is an inherent part of it.
Posted on Reply
#117
Vayra86
medi01 said:

Nice (emphasis mine).
Indeed
There is no "being mad" here.
Indeed



If you're not mad, stop swearing. Beyond that, its clear you're just looking for things to disagree on, have fun with that.
Posted on Reply
#118
medi01
Vayra86 said:

If you're not mad, stop swearing
The post to which you replied with stinky "mad" comment doesn't contain a single uncensored word, besides one could be using them without "being mad".
Why did yo react like that? BH? Sigh.

And for the other part, perhaps RTX cores are not as "specialized" as someone wants us to believe.
Posted on Reply
#119
Vayra86
medi01 said:

The post to which you replied with stinky "mad" comment doesn't contain a single uncensored word, besides one could be using them without "being mad".
Why did yo react like that? BH? Sigh.

And for the other part, perhaps RTX cores are not as "specialized" as someone wants us to believe.
Scroll through your general tone of voice on this forum and maybe the answer will reveal itself to you. Nice way to deflect the core of your incorrect statements of the last few pages though. I'm done with you.
Posted on Reply
#120
medi01
Vayra86 said:

Scroll through your general tone of voice on this forum and maybe the answer will reveal itself to you
You complained about "mad" comment, and "swearing" in the context of a post that had neither nor.
It is your misguided interpretation of the world around you multiplied by some tribalism more than anything.

I don't have any feelings (in either direction) about anyone posting on this forum.

Vayra86 said:

I'm done with you.
I don't think I care enough to remember you tomorrow. So, uh, well, just don't reply.
Posted on Reply
#121
londiste
medi01 said:
And for the other part, perhaps RTX cores are not as "specialized" as someone wants us to believe.
RT Cores do BVH Traversal and intersection testing. Specialized in what they do but what they do are operations useful to practically all raytracing implementations.
Posted on Reply
#122
ShurikN
I'm really liking the look of this Taichi prototype (minus the RGB)

Posted on Reply
#123
bug
londiste said:

RT Cores do BVH Traversal and intersection testing. Specialized in what they do but what they do are operations useful to practically all raytracing implementations.
Let him be. He's grasping at straws, in denial that AMD is missing on yet another GFX advancement.
Posted on Reply
#124
jabbadap
ShurikN said:

I'm really liking the look of this Taichi prototype (minus the RGB)

[ATTACH alt="taichi.jpg"]123927[/ATTACH]
Spinning ring of fire, or is my eyes deceiving me? Tad long card though, that size of cooling could be on Radeon VII class TDP product.

Are they finally doing their own design though, their RX 500 -series and Vegas were mostly subpar products.
Posted on Reply
#125
Vindicator
FordGT90Concept said:

It was announced in January, too late to put into Navi. Arcturus might have it.

I want to know how many transistors it has.
https://www.anandtech.com/show/11003/hdmi-21-announced-8kp60-48gbps-cable
It was announced publicly in January 2017, nearly 2½ years ago, and that's to the public. Who knows how long it was developed and discussed behind the scenes prior to this. I am extremely bummed HDMI 2.1 is apparently not supported by these cards. I expected the PCIE 4.0 announcement to be the perfect time for AMD to really jump ahead with their GPU feature support. Such a missed opportunity imo.
Posted on Reply
Add your own comment