• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Polaris architecture – GCN 4.0

Depends lot on the exact definition of mass production and the exact time frame.


Assuming mass production means there is going to be no stock constraints and micro can ship crate loads of chips to nvidia/AMD by May then for a end June released GPU there is no issue.

Mass production doesn't mean there was no production before hand, just highly constrained small batch production a few thousand chips say.

Also, mass production start date may be simple delayed until a big order is due, e.g. Micron might be able to have mass produced 4 months ago but AMD/Nvidia only need chips this May-June so Micron don't produce until then, there is no other customer so no need to mass produce early.

Wow, first up Micron have begun the process of moving to mass production, it takes time and they only got samples back a month or so ago, no they couldn't have done it earlier and it wasn't delayed.

Second, if they started shipping mass quantities in May... they couldn't release cards in June, not a chance in hell, you're dramatically underestimating the time scales involved with shipping, packaging, testing and reshipping finished products around the world.

Then you have, you know why Micron didn't announce entering mass shipping in summer, because there is actually a word available in the english language to describe exactly what you're suggesting and they didn't use it. So do Micron not know the word shipping, or did they announce when mass production would start because they meant that is when mass production would start.

Lastly.... in all this wild speculating May is now summer.

The industry at large never, ever says "entering mass production" to mean shipping final product, ever, in 50 years of the industry the term is known exactly what it means.

It's extremely unlikely we'd see any graphics cards with gddr5x in it before late Q3, probably Q4, maybe next year. Also AMD and Nvidia aren't the only customers for it.
 
Wonder why they are doing DVI again, they stopped that with Fury.

Lower end cards are dramatically more likely to be bought by those on lower budgets. So they will be significantly more likely to have an older screen with DVI, cutting them out by not including it would be madness. On the opposite end of the scale people buying a Fury X or 980ti are significantly more likely to have a modern screen with display port/*-sync.

Budget cards not accounting for older screen connections can and will lose you millions and millions of sales. Every guy who bought say a Q6600 PC from Dell that has a dvi screen and they want to buy a low power cheap graphics card to upgrade would either buy a screen and a DP only low end card, or a Nvidia low end card.
 
The fact that the new manufacturing process is nearly half the size of the one the Nano is on?

High end card, 600mm^2, power circuitry and 16 memory chips = big pcb, Fifi/HBM allows you to shrink the circuitry and remove all the memory traces.

A low end card will have 4-8 memory chips, 1/4 of the power circuitry, the actual chip and package itself will be around 1/3rd the total area and far few memory traces being a usually 128bit bus and less memory chips.

Low end cards were always much smaller than high end cards. HBM enabled a small card for a high end gpu. A HBM enabled low end gpu would be even smaller.
 
Wonder why they are doing DVI again, they stopped that with Fury.

Initially I was annoyed, considering FreeSync only works via DisplayPort (and HDMI when those monitors get released), but...

Lower end cards are dramatically more likely to be bought by those on lower budgets. So they will be significantly more likely to have an older screen with DVI

This is Polaris 10 right? The same card that appears to beat the Fury X on Hitman DX12 benchmarks? Are you trying to imply that this Polaris 10 is a cheaper card? It sounds a bit too good to be true... and I don't think many 1440p+ monitors have DVI these days. The FreeSync monitors have only DisplayPort and HDMI.

I also noticed from the site with the images, this card sports 3 Displayport 1.3 (supports up to 4k@120Hz, though DP 1.4 was just published this month) and a HDMI 2.0 (supports up to 4k@60Hz) in addition to the ununsual inclusion of DVI. What are they trying to suggest with the DP 1.3... Is this card going to be capable of high framerates at 4k or something? Again, sounds to good to be true.

If Polaris 10 is going to be a cheaper mid-ranged product, then wouldn't that mean Vega is actually not too far away, since AMD will need something to fill the top-of-the-range slot.
 
I am confused, does that DVI connector is stopping you from purchasing Polaris card? Does it stop AMD to enable all those DP ports and HDMI? Is that DVI the only connector available? I think the answer is no, so why moan about it? If you are not using it, there will be people who use it.
 
I am confused, does that DVI connector is stopping you from purchasing Polaris card? Does it stop AMD to enable all those DP ports and HDMI? Is that DVI the only connector available? I think the answer is no, so why moan about it? If you are not using it, there will be people who use it.

TBH i think it is a good indication that AMD are aiming it at the Mid - high segment. And i am quite glad that they are using full sized DP Ports. Far more robust than the mini ports.
 
Hold on, the Fury cards got blasted for not having HDMI2.0 and now the Polaris cards are getting blasted because they do have a DVI port. I'm all for moaning when either side does something wrong, but surely including 3 display ports, HDMI and a DVI isn't a bad thing.
 
Does make me laugh this forum sometimes. Nvidia have a DVI on the 980ti, no one cares. AMD stick one on Polaris and people care.

Nvidia release cards with 4gb (and 3.5gb) and no one cares, AMD release cards with 4gb and the whole world implodes.

Why is this?
 
Does make me laugh this forum sometimes. Nvidia have a DVI on the 980ti, no one cares. AMD stick one on Polaris and people care.

Nvidia release cards with 4gb (and 3.5gb) and no one cares, AMD release cards with 4gb and the whole world implodes.

Why is this?

Even though I do agree that this forum can be really fickle, I'm not sure about the no one careing about the NVidia 3.5GB issue. ;)
 

Maybe they lied about 10 being GDDR5 then ... that or they found a way to dramatically simply the circuitry and cram all the chips into an improbably small PCB. That case is very small.

Edit - if it's **Nano** sized and has more than 4GB of RAM (which it'd need for the bandwidth with GDDR5), there is no way it it is not HBM(1).
 
Back
Top Bottom