Ominous ve.ga Website Surfaces with a Word Cloud

[btarunr]

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:

• 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.

View at TechPowerUp Main Site

 

[ShurikN]

The site currently has a count down timer. Less than 3 days to go.

 

[P4-630]

 

[HD64G]

If Zen and Vega live up to their logical hype derived from leaks (ie Ryzen close to 6950 at much lower price and overclocks well, Vega clearly wins over 1080 with better vfm, etc) , AMD has a bright future. They will be succesful even if they get very close to the hype as long as their marketing team keeps hype under control.

 

[xkm1948]

I love my FuryX but damn I want this new piece of AMD toy NOW!

 

[64K]

What I want to see is the price and a thorough review done here. I wonder if any review samples have been sent out yet.

 

[xkm1948]

What I want to see is the price and a thorough review done here. I wonder if any review samples have been sent out yet.

More like whether they have started mass production yet. This hype is good entertainment though.

 

[the54thvoid]

If Zen and Vega live up to their logical hype derived from leaks (ie Ryzen close to 6950 at much lower price and overclocks well, Vega clearly wins over 1080 with better vfm, etc) , AMD has a bright future. They will be succesful even if they get very close to the hype as long as their marketing team keeps hype under control.

The bit I bolded is so unclear yet.

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.

 

[kruk]

• 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."

I really doubt this number refers to Fiji, because Fury cards are already at Maxwell's level. It must be the baseline which was used for Polaris 2.5 Perf/Watt -> Tonga. This means they will achieve around the same efficiency as current Pascal cards.

 

[ShurikN]

Did you guys notice this in the video on the ve.ga page

 

[the54thvoid]

Did you guys notice this in the video on the ve.ga page

So Vega is better than Titan X because if it's not, Vega will be dumping eggs on AMD's face.

 

[HD64G]

The bit I bolded is so unclear yet.

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.

Allow me to differ a bit. Vega 10 just needs to win 1080 in DX11 and get close to Ti in DX12 for about 1080's price. That would make it succesful. And customers need vfm as most of the ones who spend $1000 for a GPU, they would prefer that specific one supposed it would sit right where I described performance wise as Ti. If for $600 you get a GPU between 1080 and Ti on average that should be sold for $800 with just 10% more FPS, who would you buy personally?

 

[the54thvoid]

Allow me to differ a bit. Vega 10 just needs to win 1080 in DX11 and get close to Ti in DX12 for about 1080's price. That would make it succesful. And customers need vfm as most of the ones who spend $1000 for a GPU, they would prefer that specific one supposed it would sit right where I described performance wise as Ti. If for $600 you get a GPU between 1080 and Ti on average that should be sold for $800 with just 10% more FPS, who would you buy personally?

I'm one of those morons that waits for the chips to fall (these chips being Vega and 1080ti) and I buy the fastest. So, like I say, if Vega is between ti and 'non' ti, I will buy a Titan X.

 

[TheGuruStud]

I'm one of those morons that waits for the chips to fall (these chips being Vega and 1080ti) and I buy the fastest. So, like I say, if Vega is between ti and 'non' ti, I will buy a Titan X.

Unfortunately, it won't be worth the cash over the 1080ti as it can only be marginally faster. A couple hundred more cores and extra ram isn't going to do anything. That's a jerk off increase that can be made up for in clock speed on the 1080ti.

I doubt they're making new chips on 14nm. That would cost more money LOL.

 

[dj-electric]

People who get too hyped need to remind themselves that only a short while ago AMD preferred using TITAN X to showcase CPU gaming performance.

They didn't even use their own Vega early samples

 

[kruk]

People who get too hyped need to remind themselves that only a short while ago AMD preferred using TITAN X to showcase CPU gaming performance.

They didn't even use their own Vega early samples

Why would they? If something would go wrong with the GPU drivers (stutter for example) it would put Ryzen in unnecessary bad light. It was a really smart move!

 

[Blueberries]

Volta will be at CES too guys, don't get too excited yet, this may be a repeat of Polaris.

People who get too hyped need to remind themselves that only a short while ago AMD preferred using TITAN X to showcase CPU gaming performance.

They didn't even use their own Vega early samples

They did that intentionally so that people couldn't argue it was optimized for their hardware only.

I love my FuryX but damn I want this new piece of AMD toy NOW!

The FuryX will go down as one of the best card designs in history. I don't care if it soaks up as much as 300W and is outmatched by Pascal, it's a beautiful card that anyone would love to own. The CM pump with the flattened copper piping to cool the VRM is one of the most innovative things I've seen in reference board design.

Let's hope they do something equally as great with Vega and deliver on their promises of HBM2.

 

[Xzibit]

Volta will be at CES too guys, don't get too excited yet, this may be a repeat of Polaris.

Maybe in Auto products. There only 2 display booths are in Auto Tech.

 

[Blueberries]

Maybe in Auto products. There only 2 display booths are in Auto Tech.

I would be very surprised if Jen-Hsun lets AMD steal the show at CES, even if there's no official word yet, but I would bet money he shows off the goods at some point during his key note speech.

 

[Xzibit]

I would be very surprised if Jen-Hsun lets AMD steal the show at CES, even if there's no official word yet, but I would bet money he shows off the goods at some point during his key note speech.

He could always pull a CES 2016.

Nvidia Used Maxwell GPUs Masquerading As Pascal In Drive PX2 Demo

Did NVIDIA show Maxwell instead of Pascal?

NVIDIA was just not ready to show Pascal, but they failed to inform us that the Drive PX 2 prototype had Maxwell installed.

 

[efikkan]

"4× Power Efficiency" and "2× Peak throughput per clock"
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.

 

[FordGT90Concept]

AMD has first dibs on HBM2 memory. That pretty much prohibits NVIDIA from launching a competing product first.

This architecture sounds like a pretty major improvement over Polaris even.

 

[Xzibit]

AMD has first dibs on HBM2 memory. That pretty much prohibits NVIDIA from launching a competing product first.

This architecture sounds like a pretty major improvement over Polaris even.

No, Nvidia was first with HBM 2.0 both in HPC and in server market.

 

[azdesign]

The whole video in ve.ga is a cringe train lol. Hope they're still selling cards instead of drums.
Anyway,
I've been holding back from buying 1070 since long ago and hopefully amd makes the wait worthwhile.

 

[Divide Overflow]

"Make some noise"? I hope that doesn't allude to their stock cooler acoustic performance!