AMD Big Navi Would Yield 40-50 percent more than the GeForce RTX 2080 Ti

Clusters? Is this MCM (multi chip module) or "glued together" with something like infinity fabric? some sort of internal crossfire? wonder how bad the microstutter will be on this

 

If AMD can produce a similar "wonder" to Ryzen, but this time on the GPU market - this will be a major win for everyone.

MCM is the clear way to go forward.

 

Will certainly be a beast if true. AMD back in the enthusiast level category.

 

Indeed it would be, but there has been no indication that either AMD or NVIDIA have achieved that yet (both are researching it). It would also be more likely for a full new architecture instead of an "upscale" like "Big Navi".

 

One way to find out.

 

Lisa said they will compete with nvidia at high end with rdna2. I really do hope they pull out a ryzen in their gpu segment.

 

Same, our wallets really need it as its looking more and more expensive moving forward with GPU technology.

 

+1 to all said above me, I really, really hope/wish this to be true!

 

I share the sentiment, however the power consumption does look a bit scary. Might require 3 8-pins.

 

Nothing compared to the latest Ampere rumours, 600w+, new 12 Pin power connector, which means a new PSU will have to be made and bought, as there is no 12 Pin connectors

 

CPU bottlenecks everywhere...

 

And the games (more or less) shows similar GFX
There is hope now in the Next-Gen (RDNA2 & NVMe) to see some jaw dropping graphics.

 

It would be very efficient GPU if GPU was only made for raytracing.

While there were some hints that MCM for GPUs that are used for current rendering techniques may not get as big penalty as was initial expectation, it really depends on way GPUs are cut into parts. (There is need to minimize number of traces within interposer.)

No more jaw dropping than megascans of real world produce. There have been demos that are nearly indistinguishable from reality made on UE4/Frost... for years.
RDNA2 will add cycle efficient way to do DX-R, but no extra magic instructions.

It is always about ability of artist to use available computational power. And when certain "smart" techniques are used, wonderful graphics can be produced even on GCN GPUs.

RDNA2 is way forward. It will likely improve on utilization and achievable fps under conditions which gave GCN trouble even more than RDNA1 improvement.

 

I really doubt they'd make a single card that requires 600W+ for the consumer market. Maybe for servers/datacentres?

 

Top performance cards always pushed 300W TDP. Two 8 pin connectors plus the motherboard can supply total 375W, as long as they don't push over 310 I think its safe. No information has been given, but consoles are using a smaller version of RDNA2. Will be interesting to see if this is a monolith die or a surprise MMC (doubt). Can't wait for MMC to arrive, if Ryzen has teach us anything is that's going to be the future.
Nvidia has focused on making faster products, because people still pay for them. Someone has to remember not everyone can afford 1000€ of a GPU, when many still build full gaming PCs for 500€!

 

But the 2080ti already pulls ~300W, why would big Navi scare you?

 

If this is true, and AMD comes in at an outstanding price point to undercut Nvidi's 3000 offerings, you may want to think about selling your 2080ti's now. No one is going to pay top dollar for them if Big Navi comes in at a good price point.

 

No it doesnt.
https://www.guru3d.com/articles_pages/geforce_rtx_2080_ti_founders_review,9.html

 

Im down for both, i'd buy amd or nvidia. Whoever releases first, i'll hit it. Im growing sick of my 2080 which barely handles RDR2 at decent tweaked settings but not as high as i would like.