Monday, January 2nd 2017
Ominous ve.ga Website Surfaces with a Word Cloud
An ominously named website ve.ga surfaced on the web, pointing to the placeholder of a website AMD dedicated to its upcoming "Vega" GPU architecture. This placeholder has a word-cloud (or tag cloud), which is a 3D spacial clump of keywords/tags sorted by popularity. The words in this word-cloud could spell out key features of the "Vega" architecture, which AMD is expected to unveil later this week.
Here's what we make of these keywords:
Source:
3DCenter.org
- 8x Capacity/stack: This probably refers to the fact that HBM2 memory has 8 times the capacity per DRAM stack, as the first-generation HBM. While HBM1 maxed out at 4 GB over a 4096-bit interface (1 GB per stack), on the Radeon R9 Fury series; HBM2 allows you to cram up to 32 GB of memory (or 8 GB per stack), over a 4096-bit interface. This also means that whatever Vega-based AMD is going to launch, will come with HBM2 memory.
- 2x Bandwidth Per Pin: Refers to the increase in bandwidth per pin of HBM2. With a memory clock of 500 MHz, 4096-bit HBM2 could reach 1 TB/s. This gives AMD the cost-saving flexibility of deploying a 2048-bit memory bus with 8 GB of memory, and yet ending up with 512 GB/s bandwidth (more than that of the TITAN X Pascal).
- High Bandwidth Cache: This is a very curious phrase. Perhaps it means that AMD managed to increase the data throughput of the GPU's on-die caches, or, given the way this is worded, it could be a larger cache faster than the video-memory, which the GPU can use as a large scratchpad. The fact that another keyword in this cloud refers to a "high-bandwidth cache controller" adds weight to this theory.
- 512 TB Virtual Address Space: This adds even more weight to the theory that the HBM2 video memory, and the on-die L2 cache may not be the only physical memory on "Vega" graphics cards. The GPU probably needs a vast address space of up to 512 TB, to juggle data between the VRAM, caches, and the host.
- 4x Power Efficiency: This is probably just a 4x performance/Watt increase over the "Fiji" silicon, and shouldn't come as a surprise, given the new 14 nm process, and the energy-efficiency improvements AMD already made with "Polaris."
- Next Generation Compute[r] Engine: This probably refers to an upgraded compute unit (CU) design, which is at the heart of the 4x performance/Watt increases
- Primitive Shaders: If this is what we think it is, it's big. Most modern shader operations can be simplified into smaller, simpler operations.
43 Comments on Ominous ve.ga Website Surfaces with a Word Cloud
If Vega is better than 1080 (and it BLOODY should be) then AMD will probably price it accordingly. I think people forget what AMD did with the 7970 and Fury X. They matched the contemporary high end prices. Also, we actually need Vega to be as fast as the Titan X. With the same shader count as Fiji, on 14nm, with HBM and a newer architecture, using those primitive shaders (which i know nothing of) it has to be better than speed bumped top end Maxwell (AKA GP102). People should be disappointed if this isn't better than Titan X. If it isn't, then GTX 1080ti will simply brush aside of all this wonderful, boastful AMD hype and push it into the dirt.
Now lets make it clear, I want Vega to be faster than Titan X. I want better competition. I want Vega at 1080 prices that matches Titan X performance. Given how competent Fiji was (and is), you'd expect RTG to nail this one. But if Vega just matches the 1080 (not Nvidia's top tier chip, not even close) that's a super fail.
I doubt they're making new chips on 14nm. That would cost more money LOL.
They didn't even use their own Vega early samples
Let's hope they do something equally as great with Vega and deliver on their promises of HBM2.
Nvidia Used Maxwell GPUs Masquerading As Pascal In Drive PX2 Demo
Did NVIDIA show Maxwell instead of Pascal?
I guess this is referring to the fp16 support doubling the throughput of fp32.
If they still managed to deliver 2× power efficiency for fp32, then it would put them ahead of Nvidia. That would be awesome and a giant comeback, but keep in mind they promised 2.5× power efficiency with Polaris and never delivered anything close to that.
This architecture sounds like a pretty major improvement over Polaris even.
Anyway,
I've been holding back from buying 1070 since long ago and hopefully amd makes the wait worthwhile.
The specs on this thing do look great. But if the teasers were AMD showcasing their top tier model and it only beats a 1080 in Doom running Vulkan, that doesnt inspire confidence.
There is a heck of a gap between the RX480 and a GTX1080 tho, so something that generally comes out on top of the 1080 if priced correctly would still have a place.
But my next GPU will be whatever is the best for gaming that doesnt cross the grand marker.