Discussion in 'Videocards - AMD Radeon' started by WhiteLightning, Sep 28, 2018.
That makes a lot more sense, fingers crossed!
For Us Vega owners this is not an upgrade
Too much of a good Features already in Vega 1st & 2nd Gen respectively.
HBCC on HBM2
ACE's at 16x and many more.
Not long now, 27.V
Bold, moment you pick particular metric, it becomes subjective matter. "Care" For "A" is opinion since there is B, C, D, ... and in your opinion "A" is important. And personal preferences... how objective.
Underlined: Crystal ball.
Again, your asking AMD to quit after not being the best would lead to never having actually marketable GPU architecture for consoles in 1st place. Hell, situation ATi was in at time AMD acquired them was so bad that you would have never got them in 1st place, right?
And yes, if AMD did quit this market as you did suggest, it would be exactly that. Who else would remain in PC GPU market outside nVidia? Who else would remain in PC CPU market outside intel?
You make logical contradictions almost on every step you take. And thanks for not writing more of it.
What is the lowest Navi planned, how much could it cost and when would it be released? I know those are speculative questions but I don't read all news so I am bit uninformed.
Demonstratively it look like this.
Lowest model will be PCIe only no add. Pin and price shold be less than 100€
Wait until 27 may for more substantial info form ATI/AMD
RX 640 seems to be rebranded Polaris based RX 550X with 512SP, 32TMUs, 16ROPs. (But it could as well be 640/40/16 variant.)
Therefore Lowest Navi should have more than 512/640SP or be able to deliver higher performance through higher clock. I would guess that lowest will have 1024 or 1280SP. And cost around ~$100. Basically being just under RX 570 in terms of performance.
Thing is that AMD's Polaris range is pretty crushed together at this moment. And deciding what price to put on what performance is pretty damn hard. 896SP RX 560 price is just $10 from RX 570 which has 2048SP (and is considerably stronger + is bundled with few valuable games... basically making it quite cheaper).
Depending on your needs/budget, it may not be bad idea to get RX 570/580 as Navi drops in. (Or maybe even before as Games may no longer be bundled upon release of Navi.)
I'll add my theory on Low-end Navi:
- Navi is made to work with GDDR6 and HBM2
- Neither of those is cheaper per GB than GDDR5
Therefore pairing super weak GPU with GDDR6 will make card too expensive for being weak and wasting GDDR6 bandwidth
Lowest Navi having 4x 1GB GDDR6 and ~140~160GB/s bandwidth would still have like 40% more bandwidth than RX 560, therefore I would expect it to have performance somewhere around that. In terms of recent GPUs, it would be around GTX 1050Ti.
On other hand, there is another option:
- There are 2GB GDDR6 chips with different bandwidths
- Therefore there may be 2x 2GB configuration which could easily have 112~128GB/s bandwidth
- Paired with Navi based GPU performing just above RX-560
-> It would eat ~45W and have cheaper cooling than RX-560
-> It would have cheaper (fewer) VRMs phases to match lower power draw
-> PCB would be cheaper and smaller too as only 2 memory chips
-> GPU itself would have smaller IMC and graphical part would need fewer SP/TMU/ROPs as performance target would be reached through higher clock. (compared to same RX-560) This would offset 7nm cost making it about same.
-> Only thing more expensive would be GDDR6
But altogether even Navi can occupy RX-560 performance spot and be bit more economical there.
Oh for sure, I weren't planning to upgrade my GPU anytime soon (if I were aiming at Navi I wouldn't have bought this V56 last November).
I'm just enthusiastic about what it brings to the table, partly due to previously mentioned reasons and partly because.. uh, let's call it "general hardware enthusiasm". It's the right forum after all.
Big Navi should be next year. These price look good, maybe too good. AMD needs to profit as well as gain market size.
So will Vega 56/64 be dropped for Navi, with Vega VII being top tier?
I'm also wondering if Navi would be worthwhile for me to upgrade to or just stick to my Fury til the generation after Navi?
Vega 64 and above seems like the only option to me currently and I'd rather pay under £300. Don't think Vega 56 is compelling enough currently.
I think Navi will be worth an upgrade. The Vegas will be fazed out. Might see some really good deals.
Vega 56 have OC headroom, but if you are use 1080p the Fury is still a pretty good card.
Consider only Vega 64 Nitro+ or similar 3xFan
If You find deal at ~280-375€ take it
Or if You got lucky & find some XTX -----> no matter, just pick it up and be happy.
Vega still selling great...
Omna, what is great ?
I have very limited info ( Steam survey and following Gamers Nexus video: ), but it seems to me that AMD is not doing well at all, exept for some budget cards.
As much as I really want AMD to pick up their game again ( I was a nVidia fanboy, but after reading their busines pratises, not so much. Going underdog! )
The chart clearly shows a bit of a peak at launch and a few months after, but after that...
Let's hope next AMD gen fixes this and makes the consumer as a whole believe a bit more in AMD.
Im happy with my Vega (Great Tech & very powerfull)
There is no struggle, all in all ATI need good GPU up to 250€
(IMhO Navi will deliver)
AMD "Navi" Features 8 Streaming Engines, Possible ROP Count Doubling?
AMD's 7 nm "Navi 10" silicon may finally address two architectural shortcomings of its performance-segment GPUs, memory bandwidth, and render-backends (deficiency thereof).
The GPU almost certainly features a 256-bit GDDR6 memory interface, bringing about a 50-75 percent increase in memory bandwidth over "Polaris 30."
According to a sketch of the GPU's SIMD schematic put out by KOMACHI Ensaka, Navi's main number crunching machinery is spread across eight shader engines, each with five compute units (CUs).
Five CUs spread across eight shader engines, assuming each CU continues to pack 64 stream processors, works out to 2,560 stream processors on the silicon.
This arrangement is in stark contrast to the "Hawaii" silicon from 2013, which crammed 10 CUs per shader engine across four shader engines to achieve the same 2,560 SP count on the Radeon R9 290.
The "Fiji" silicon that followed "Hawaii" stuck to the 4-shader engine arrangement. Interestingly, both these chips featured four render-backends per shader engine, working out to 64 ROPs.
AMD's decision to go with 8 shader engines raises hopes for the company doubling ROP counts over "Polaris," to 64, by packing two render backends per shader engine.
AMD unveils Navi in its May 27 Computex keynote, followed by a possible early-July launch.
Capcom is excited by the PS5's loading times and VR tech
Sony's PlayStation 5 should be quite beastly. The system will feature all new customized AMD chips including a potent Navi GPU in tandem with a Zen 2 CPU to deliver high-end console gaming.
Other features like full native backward compatibility with PS4 games will ensure a smooth transition.
But Capcom also sees big opportunities with other features like the PS5's ultra-fast broadband SSD.
In a recent interview with Weekly Famitsu (as translated by Twinfinite), Capcom's games division head Jun Takeuchi says the PS5's beefy specs and blazing-fast memory can drastically improve immersion.
Loading times break immersion and action games like Devil May Cry stand to gain quite a bit from the PS5's expanded hardware, and Capcom is eager to use the console to its full potential to optimize its games.
Seamless play across all devices was a big part of Sony's new business model and Takeuchi agrees with this approach.
Sony is putting emphasis on being able to play PlayStation games anywhere regardless if you own a console or not.
Capcom is a big believer in this idea--they've experimented with streaming their games in Japan on the Switch--and wants to diversify its offerings using cloud-based services.
Sony strongly hinted PlayStation Now will come to mobile phones at some point and when it does Capcom should be a big part of this push.
AMD Navi RX 3080 Specs Leaked
The countdown to AMD’s upcoming next generation Zen 2 & Navi announcement on the 27th of May is quickly ticking down.
We’re now less than four days away and the leaks just keep on coming.
Today’s leak pertains to the company’s upcoming mid-range Navi GPU, which some sources pin as Navi 10 whilst others claim is actually called Navi 12.
Even though the company’s next generation nomenclature is still uncofirmed, for the purposes of brevity and simplicity we’re going to label this chip Navi 10 and the highest specced graphics card it powers the RX 3080.
AMD Radeon Navi RX 3080 Specs Leaked – 8 Shader Engines, 2560 SPs, 8GB vRAM & 256-bit Bus
Before we get into the juicy bits as it were, we should first point out that this leak is courtesy of venerable leaker KOMACHI, although his tweet has since been taken down we’ve managed to grab a hold of all the tasty bits.
Before we proceed it goes without saying that you should definitely take this upcoming information with a grain of salt, as always nothing is confirmed until we actually have the hardware in our hands.
So without any further ado, let’s dig in.
Navi Allegedly Brings New Shader Engine Design For Better Compute Unit Utilization Efficiency & Higher Pixel Throughput
AMD Navi 10 Diagram by KOMACHI
According to this leak, as well as our own sources, the Navi chip features 40 compute units, each housing 64 stream processors.
The new previously unknown detail that this leak brings to the table is that Navi allegedly features 8 shader engines.
Each of these engines will house 5 compute units and a raster back-end. Normally, in existing GCN GPUs, a raster back-end would include 16 Render Output Units, or ROPs for short. So it’s unclear if Navi 10 will feature 128 ROPs or if the company is cutting down the number of ROPs per render back-end with Navi.
Now, the number of sahder engines in Navi is actually a significant detail because one of the major architectural limitations of AMD’s previous GCN implementations is the inability to scale the microarchitecture beyond 4 shader engines per die.
This has led to underutilization of stream processors in the company’s bigger GPUs such as Fiji and Vega and has long been a known limitation of GCN that the company was working to develop solutions to.
AMD Kicking Things Off With Mid-Size Navi, Bigger Chip to Come
If this leak is accurate, AMD’s engineers have successfully overcome this limitation and as a result we should see better performance scaling with the company’s upcoming big chips compared to its previous designs.
This indicates that we could very well see new Radeon GPUs with an excess of 4096 stream processors, which has been the maximum we’ve seen since the introduction of Fiji four years ago.
In theory, the expansion of shader engines would also allow the company to add more than 64 ROPs to its future designs which would result in significantly improved pixel throughput, another limitation of its previous big chips.
Two Navi's 88 in CF sounds nice, is there a chance or are they gonna dropp it for good?
If Vega2 would offer it, i had def. tried it
it's happening (Live now)
RDNA is Navi (New uArch.)
HD4xxx & HD5xxx Era strikes again
Vega64 + 15% for RDNA 57xx (there will be 58xx & 59xx also)
AMD Launches RDNA Macro Architecture: PCIe 4.0, 1.25x IPC And 1.5x Performance Per Watt Upgrade Over GCN
The AMD RDNA (RDNA is a clean-slate GPU design that is not based on Graphics CoreNext) architecture will feature 1.25x times the IPC and 1.5 times the performance per watt. Most interestingly however, it features a completely new compute unit design.
This means my exclusives on Navi may need a revision – since all of them were based on the CUs having the same CU-to-SP count ratio as GCN.
The new compute design features improved efficiency as well as a new multi-level cache hierarchy. This results in reduced latency, higher bandwidth availability ( on a per-core basis) and lower power consumption.
The RDNA macro-architecture also features streamlined graphics pipeline which features “high clock speed”.
This might hint to the fact that the company is going to be chasing after high clock speeds seen in its competitors product which can clock up to 1900 MHz (in comparison, AMD GPUs usually stick around 1500 MHz).
If this is true then we should be seeing some pretty big performance gains.
The all-new RDNA macro architecture is expected to stick around for a while.
To those who are wondering, the Next-Gen architecture on AMD’s roadmap will be a micro-architecture under RDNA and the Next-Gen is just a place holder.
Navi architecture is the true successor that we have been waiting for a very long time.
It also overcomes one of the biggest limitations fo GCN architecture, and the real reason AMD was pushing for this: the 4096 SP limit that was holding back Radeon Graphics.
Finally, here is the kicker, RDNA fully supports PCIe 4.0 which means NAVI GPU is slated to be the first PCIe 4.0 gaming GPU in the world.
This is great news because it means the company is looking to future proof its lineup for the get go – something its known for and has a considerable amount of goodwill for.
All in all, not much other information was revealed about the architecture and we can’t wait to hear more details at E3. Exciting stuff is coming.