The last times I was asked I used the GTX 470, and then a GTX 980.
None of my AMD/ATI cards were ever listed, never mind all my other NVIDIA cards.
It's a silly thing.
-
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 VEGA confirmed for 2017 H1
- Thread starter Doom112
- Start date
- Status
The last times I was asked I used the GTX 470, and then a GTX 980.
None of my AMD/ATI cards were ever listed, never mind all my other NVIDIA cards.
It's a silly thing.
I actually think if it was 1080 + 20% it might be closer to 600.
It can't be cheaper to produce an HBM2 card rather than a GDDR5/X card?
All the R&D has to be taken into account.
Can't get a quote for HBM would need to design an entire interposer for fabrication apparently - GDDR5 is about $2/chip for small quantities and $0.5/chip for 5000+ (no idea what the cost would be to nVidia or AMD) the rep reckons in comparison it would be $14-60/chip for HBM type stuff but that might not be accurate.
EDIT: That doesn't include things like the cost of the controllers, etc.
Last edited:
It just chance, just becausr you weren't asked to re-take the survey doesn't mean there is any bias.
They only have to ask a few thousand users to get very solid statistical data, whether you have never bean asked or get asked multiple times makes no difference.
its like rolling a dice 6 times. Some users will never see the number 6, others might see it 6 times but the dice itself is truly random. Anecdotes of someone rolling a dice 4 times in a roll doesn't change the statistical outcomes.
has extra cost due to the interposer also.
HBM3 and such be cheaper and may reach similar price points but GDDR has been around a long time so that cost has been driven way down. Ram used to cost 200euro for 4gb once and now its like 40euro.
a new HDR 144hz screen is offered at 2000euro at 4k, so new tech will cost an arm and leg until the infrastructure is there.
4k was expensive also at first but cost went down fast as the customer was ready for them.
main advantage with HBM is the exchange of memory swap on the fly which GDDR cant do. Vega will have new tech and I know I dont buy old tech myself.
AMD also have access to GDDR5X, so the fact that they choose HBM2 means it's worth the additional cost, performance/heat/power wise.
Also bear in mind that the Vega 10 part shown has only 2 stacks of HBM2. Fury had 4 - so there's already half the complexity regarding the interposer and TSV's, as well as other components. All of which mean the memory solution used on Vega will be far cheaper than the one used on Fiji.
Also bear in mind that the Vega 10 part shown has only 2 stacks of HBM2. Fury had 4 - so there's already half the complexity regarding the interposer and TSV's, as well as other components. All of which mean the memory solution used on Vega will be far cheaper than the one used on Fiji.
Steam survey is a load of rubbish. I have never been asked to take part in the Steam survey and I've changed 5 or 6 cards since I started using it. I consider the survey results meaningless since none of my cards have been counted.
Steam could easily collect user hardware info automatically but they don't and appear to be very selective. AMD have gained more than 8% market share since Polaris release yet steam shows the 1080 with more owners. Are we to believe more 1080's have sold than the RX480?
Last edited:
People don't like the idea of automatic telemetry - most companies avoid doing it unless they are MS or nVidia :s
Dunno why people are arguing about it - its a purely random polling of a cross section of the user base large enough to be representative of the whole nothing more or less.
Dunno why people are arguing about it - its a purely random polling of a cross section of the user base large enough to be representative of the whole nothing more or less.
Well the HBM didn't seem to be worth it on the Fury.
I think AMD will be still trying to recoup the substantial costs to develop HBM cards due to the Fury not selling that well.
I think one of the major problems with gaming these days is the number of people reinventing the wheel.
Each company having made not just one 3D engine, but sometimes several.
What's needed really is a single engine that everybody uses, that is as efficient as humanly possible, using low-level APIs, assembler, whatever. And everybody builds their game on this engine using the engine's much higher level API/script.
Instead everybody is developing (still) their own engines, year after year after year.
Think of all the man-hours being spent duplicating that effort!
But then I tend to think of 3D engines less as competing products and more like a solid mathematical foundation on which to build. Which is probably why I'm not in business
Hard to answer as what I may consider failure others may call success.
If their best card is only 1080 performance, that is a big failure right there for me.
Yeah, given that a 1080 OC and a 980ti OC aren't that different, if Vega only (barely) matches a 1080, then they've spent the last *two years* just catching up to nVidia of two years ago.
Which would leave the ball firmly in nVidia's court.
The main reason why people doubt AMD here I suspect, is because AMD always talk big, but their talk and what they release has no correlation at all. They always talk big, but have released some real turds. Which isn't to say they don't get it right, sometimes.
Just that you can't infer anything from AMD talking the talk. They always do, regardless.
I can't remember submitting mine, friends are the same.
I wouldn't even argue the point with the guy. If he wants to rely on a random inconsistent survey for insight, let him. Not worth wasting your time
.I can't remember submitting mine, friends are the same.
I wouldn't even argue the point with the guy. If he wants to rely on a random inconsistent survey for insight, let him. Not worth wasting your time
LOl, it's not random and it's not inconsistent.quite the opposite, it accurately correlates with all other measures of sales and marketing share,which is why it is widely used in industry by game developers to target feature sets
I think one of the major problems with gaming these days is the number of people reinventing the wheel.
Each company having made not just one 3D engine, but sometimes several.
What's needed really is a single engine that everybody uses, that is as efficient as humanly possible, using low-level APIs, assembler, whatever. And everybody builds their game on this engine using the engine's much higher level API/script.
Instead everybody is developing (still) their own engines, year after year after year.
Think of all the man-hours being spent duplicating that effort!
But then I tend to think of 3D engines less as competing products and more like a solid mathematical foundation on which to build. Which is probably why I'm not in business
Doesn't work that way. No free lunch.
What you are saying g is like there should only be 1 design if car for use in racing,school drop off, off-road information, towing heavy loads, carrying many people, etc.
As with everything, specialisation is the key to performance.
There aren't many things to specialise in.
Vast outdoor areas vs indoor, tight corridors, sure.
Other than that, how many FPS games could swap engines without the player really noticing a difference.
Are you saying each company needs to make its own 3D engine, in an ideal world? Or even that each game needs its own engine? Surely not.
Random in the times people have been asked due to information in this thread and my own personal experience. Inconsistency is in relation to having several cards, both AMD and Nvidia that Steam know nothing about due to participation being selective. I understand what you're saying, the survey is working as intended but if you want to use that as concrete evidence, or even anywhere near to, you're not going to get that from Steam as your source due to reasons mentioned.
There's not much of an argument here. I'm not saying I know the divide between users but I do know that I wouldn't be using Steam for a source of enlightenment.
Fury and Vega are different cards.
Also stop to think what the Fury cards would have been like if they had regular GDDR5. The power consumption would have been way higher, the PCB would have been much larger, etc etc.
In case I'm not being clear, take a Titan XP and put 12GB HBM or HBM2 on it, it will run faster, consume less power and produce less heat.
Just because Fiji (Fury) cards were absolutely terrible doesn't mean HBM is.
- Status