This tells me that you actually have zero idea what Infinity Fabric is, and what it does.
Please, stop posting in this manner.
This tells me that you actually have zero idea what Infinity Fabric is, and what it does.Simple exercise:
How many "pins" are involved with SLI/Xfire ribbon cables.
Now let your imagination run wild with fabric.
If you don't see it, this Holiday season, you will.
Which sounds like what they plan to do with small Vega. AMD requires better thermal and electrical efficiency to break into more mobile designs. It appears to me that Vega will dramatically improve these metrics which will be wonderful if a small Vega shows up around Polaris 10's die size.Or, conversely of course, to feed the improvements from Vega back to a polaris sized chip & launch a whole new generation that way.
How about the exact TFLOPS, mem config (2 stacks)? DP being 1/16 is already confirmed I believe. And Vega 20 on a new 14nm node makes perfect sense, since they still need a 1/2 DP GPU to replace the aging Hawaii.No, node plans do not change that shortly before release, since node is one of the first things that are locked down when planning a new design (not to mention that Vega 20 on 14nm makes zero sense, since then it would just be identical to Vega 10). At best the slides were correct about Vega 20 on 7nm, but Vega 20 has since then been canceled altogether, or at worst the slides were fake and Vega 20 was never planned for 7nm (and quite possibly doesn't exist at all).
And what accurate info are you talking about exactly. Most of the info was either already known (64CUs, HBM2, packed math), or is still unconfirmed (DP rate, Vega 20, Vega 10 x2).
The only thing that wasn't known at the time was the inclusion of hardware page management support (aka HBCC), however it is interesting to note that this wasn't mentioned at all back in september 2016 when Videocardz first reported on the slides, instead they waited until the end of december 2016 for some reason. In other words Videocardz had to wait until AMD had already announced the existence of HBC and the HBCC, before they themselves could reveal that the slides had apparently mentioned this all along (even though Videocardz made zero mention of this back in september).
So they have the ability to have software transparent mGPU, but they need developers on board? How counter intuitive!Caspian & Cream, a lead AMD engineer talk in detail about Infinity Fabric and mentions that APU, GPU or CPU, it doesn't matter...!
Same event, lead software house mentions they are near 100% scalability.. in their own software, but more can be gained, given it's newness. Same even AMD talks about getting many of the software houses on board to their new technology & multi-rendering
Also, over the last 2 months many behind the scenes personalized conversation also took place. That hints at AMD path, which are not so different than Dr Su's key note speeches. That suggest new gaming technology. (We all know SLI/Xfire & plx chips are old tech).
So what you are saying is, that since AMD hasn't released their bombshell, you can't discuss it, or even allow oneself to admit it exists..? Or even discuss it in a Vega/Navi rumor thread..?
Hogwash..
How about the exact TFLOPS, mem config (2 stacks)? DP being 1/16 is already confirmed I believe. And Vega 20 on a new 14nm node makes perfect sense, since they still need a 1/2 DP GPU to replace the aging Hawaii.
Remember, they're talking a new 14nm, no different than how NVIDIA is moving "12nm".
I believe you are mistaken about MI25 showing up before Videocardz. I've been following Vega rumors fairly closely, and Videocardz was the first to say the TFLOPS figure. I have high confidence in the Videocardz leak. Plans might have changed, but at some point this was AMD's roadmap.Instinct MI25 showed up in the TPU database roughly a month before the videocardz article (unless google is lying to me), so that would give you the TFLOPS. Mem config is pretty damn easy to guess, it's either going to be 1, 2 or 4 stacks, 1 is too little, 4 is overkill (unless clocked low), so 2 is the obvious choice (especially since this would give you the same bandwidth as Fiji which is also 64 CUs).
DP has not yet been confirmed at 1/16, neither for Vega FE nor for MI25, so we don't know. Obviously if Vega 10 is capable of 1/2 (and possibly limited to less in Vega RX and Vega FE), then there wouldn't be any reason for Vega 20 to exist on 14nm. Also it's worth noting that whilst Hawaii was capable of 1/2 rate DP, it was limited to 1/8 in consumer cards, so even if Vega FE shows up with less than 1/2 rate DP, we would still really need to wait for MI25 to have confirmation on the DP capabilities of Vega 10.
Regarding Vega 20, I think it's also noteworthy that since Videocardz leaked it back in September, there has been zero independent leaks on it. Normally we would have gotten additional leaks and hints of it's existence by now, but we've gotten nothing (unless I missed something).
Anyway it's perfectly possible that the Videocardz leak is real, and something weird just happened with Vega 20 (canceled on 7nm, or otherwise removed from AMDs roadmap), but I'm just saying that I would take said leak with a large grain of salt.
I believe you are mistaken about MI25 showing up before Videocardz. I've been following Vega rumors fairly closely, and Videocardz was the first to say the TFLOPS figure. I have high confidence in the Videocardz leak. Plans might have changed, but at some point this was AMD's roadmap.
Well, as confident as one can be about rumors. There is always the small chance of something being false, but this feels like a true leak to me.
I still think about just getting a 1080ti. Amd can delay again for all we know.Yes, Vega can't come soon enough. I almost bought a 1080 Ti earlier than I had planned. I'll probably still get one, but I plan to pair it with a Skylake-X rig and put Vega in at least one of my AMD rigs.
If you don't see it, this Holiday season, you will.
Link one?You made the same kind of wild predictions in the Polaris thread for the 500 series refresh and were wrong across the board.
No one here has any reason to believe what you post and I don't suspect you have good reasons for believing what you claim either.
I still think about just getting a 1080ti. Amd can delay again for all we know.
Also I notice in the amd camp people simply talk about highend chips. Very few people actually bought and own high end amd chips here. It just seems like something people talk about and very few people actually buy.
Except there are far more 1080ti/ 1080 users than amd high end users. You can't deny that amd high end gpu gets a lot less love than Nvidia high end(for good reasons).Thats 99% of these boards. People talking about 1080 / 1080 Ti yet play at under 1080p with a $50 GPU.
I can see your point.According to google the MI25 was present in the TPU database from at least the 18th of August, whereas the Videocardz article went up in September. I think it's possible to mess with the dating of google search results though, so this admittedly isn't foolproof.
And yeah, it may very well be true, but given the parts that have turned out to be incorrect (Vega 20 on 7nm, although this doesn't necessarily indicate the leak is false, it could just be a case of changing plans, i.e. cancellation of Vega 20), and the fact that very little of the genuinely new and significant stuff in the leak (Vega 10x2, Vega 20) has been corroborated by independent sources, I feel like the prudent thing to do is to take a cautious approach here.
Except there are far more 1080ti/ 1080 users than amd high end users. You can't deny that amd high end gpu gets a lot less love than Nvidia high end(for good reasons).
Why is it that so few users actually go out and get an amd high end gpu since it seems there is so much excitement for their gpus each year?
Except there are far more 1080ti/ 1080 users than amd high end users. You can't deny that amd high end gpu gets a lot less love than Nvidia high end(for good reasons).
Why is it that so few users actually go out and get an amd high end gpu since it seems there is so much excitement for their gpus each year?
I have a Fury Nitro, a high-end RX480, and a heavily modified R9 290 (AC Extreme III) in my Ryzen rigs. All represent one half-step below the absolute pinnacle AMD card of each generation: Fiji, Polaris, and Hawaii.
Or, conversely of course, to feed the improvements from Vega back to a polaris sized chip & launch a whole new generation that way.
That's only true to a certain degree - the GPU in the PS4 pro is largely Polaris, but has elements from Vega in it. A lot of features are somewhat modular.You can't do that. The whole architecture needs to work in unison, you can't just add bits and pieces and have it magically work perfectly. The whole architecture has to be designed with those improvements, build around them to use them at the fullest, etc...
I think you misread what he was saying. He was merely saying a Polaris sized Vega iteration would be nice in the future, which is all but an inevitability.You can't do that. The whole architecture needs to work in unison, you can't just add bits and pieces and have it magically work perfectly. The whole architecture has to be designed with those improvements, build around them to use them at the fullest, etc...
Which again goes back to my point on Vega being so late but also so uncompetitive. GTX 1060 is roughly as fast as a 480/580. So a Polaris based GTX 1070 competitor should be as faster as a GTX 1070. But Vega is an IMPROVED Polaris. Why isn't Vega blowing everything out of the water?That's only true to a certain degree - the GPU in the PS4 pro is largely Polaris, but has elements from Vega in it. A lot of features are somewhat modular.
On the other hand, there's no doubt that AMD will be launching one or more smaller Vega dice down the line.
Hopefully we're going to see more of where Vega stands in about 3 1/2 hours ^_^