Will AMD release any high performance GPU's in Q2 or Q3?

**SoulForge** · 2017-05-09, 11:48 PM

Originally Posted by Gaidax

My sources tell me it's gonna be faster than Voodoo 1.

Are you sure? Those were screaming cards. Wish 3DFX (Glide) was still a thing.

**Thunderball** · 2017-05-10, 12:56 AM

Originally Posted by Evildeffy

The reference design cooling of a GTX 1060 is better than RX 480, that is correct.
However your figures are considerably wrong here.
And I mentioned PowerColor because of their cheap ass designs, however those that artefact on stock is not the fault of AMD's GPU but the absolute garbage PowerColor produces, and whilst it can happen, most do not artefact on stock or even overclocked, but don't push your luck too far on them.

I'm not blaming AMD for what PowerColor did, I'm just saying that Nvidia consumers have it better due to their GPUs having lower PCB engineering requirements.

Originally Posted by Evildeffy

Using liquid metal TIM to bridge certain control chips/voltage regulators is voiding the warranty, regardless .. pushing higher frequencies will exponentially increase power draw and heat generation.

Wont void the warranty unless your liquid metal compound is aggressive enough to eat through solder joints, and yes, it's practically useless for an average user, that's why I like the implementation.

Originally Posted by Evildeffy

That's a seriously shoddy argument to use for saying "RX 480 limit is memory bandwidth" ... you take the fact that there are no reference designs for it because it's a refresh as a sign of bandwidth limitation when there was 0 need to introduce a reference design as the new batch of Polaris chips are identical to the old just more matured.
Come now... that's grasping at straws, besides if that's the case then overclocking the VRAM to 9Gbps would see a bigger yield than increasing core speed.
Which is not the case, and yes you can overclock VRAM to 2250MHz tops in Wattman (God I hate that name).

Not connecting it in any way. Reference designs were axed because AMD would end up effectively competing with partner cards. Turbine, which is prefered for OEM sales, wouldnt be sufficient, designing a blower cooler is pointless.

You can overclock RAM, but the point is: 1) it's not designed to work like that, properties of VRAM made by Samsung or Micron are also substantially different 2) VRAM have same stock clocks, regardless of the card manufacturer, so there is no way of achieving a higher memory bandwidth without opting for a different memory type or increasing memory bus width.

Originally Posted by Evildeffy

But unlike DX12, Vulkan requires a SoC to be compatible with the standards set forth, whether that's hardware or software is irrelevant.
Also Apple's Fusion has no relevance to Vulkan as Apple does not want to work with Vulkan and instead chooses it's own proprietary "Metal".

That's why Qualcomm are going to certify 3 year old chips for running it? Vulkan standardizes hardware level optimizations, but they are not required to run it, and it certainly doesnt mean that anyone except for AMD is going to make those hardware level optimizations.

Originally Posted by Evildeffy

Again, if developers choose to expose all the capabilities of low level hardware than nVidia is indeed slower in DX12, this is a pure fact.
That developers choose not to use that or are in a stage where DX12's learning curve is still beyond them is a separate matter which does influence the "real world" ... that said nVidia is not faster with DX12 runtimes vs. AMD's implementation.

You mean designing their products around using the hardware level optimizations that miniscule level of users possess? Why would they do that? AMD implementation right now is foreign for developers for the same reason as why so few games are good at multithreading the workload: average consumer doesnt have that hardware, and wont have for some time, no matter what AMD does.

Originally Posted by Evildeffy

Point 1 is irrelevant to the answer, you stated a point of 1080Ti using less power for RAM than Polaris, that was refuted.
Point 2 is not based upon the actual voltages and current drained by that specific memory, basing it off the VRM specs is not the same numbers, VRAM drains it's own power.

Absolutely relevant point. 2-2.5 times more performance, about the same memory power consumption. Of course it's not going to scale perfectly, but it's Nvidia's most power hungry card atm (except Titan ofc, let's not count those here). VRM is the only place where you can measure power draw (of course it's not perfect, you have heat loss and power needed to operate the VRM, but 310+30=/=400 (which is a power draw of a whole 1080 Ti card).

Originally Posted by Evildeffy

Point 3: It would as it would decrease the power capabilities of a GPU, increasing board complexity and ... well you get the idea. Still out of nVidia or AMD's hands.

It wouldnt. Yes, it would require a more complex VRM design, but it's not something that hasnt been done before, some manufacturers still overengineer their boards for no apparent gains, this way we could put that engineering to use.

Originally Posted by Evildeffy

Again you did not see the point, you specifically stated that because AMD's cards are higher IPC they need a wider bus to accomodate it.
This is not true, they do not NEED a wider bus to accomodate it as it's fully dependant upon architecture and driver capabilities.
Colour compression is 1 of them for example, they just choose to design it that way, they could just as well have designed it with GDDR5X and would not have gotten any better core performance out of their Polaris architecture.
You for yourself hold that "fact" that AMD is memory bandwidth constrained in their architectures... which isn't the case.
Again if that were true we would gain considerably more performance overclocking VRAM than it would GPU core.

They need higher memory bandwidth, yes, wider memory bus is just means to that goal. I'm not saying that current cards bottleneck themselves because of insufficient memory bandwidth, but it probably did prevent them from making wider GPUs based on Polaris architecture. Not the only factor of course, others being die size (bigger the die lower the yield), GPU power consumption and thermals.

**Evildeffy** · 2017-05-10, 01:41 AM