• 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.

Poll: ** The AMD VEGA Thread **

On or off the hype train?

  • (off) Train has derailed

    Votes: 207 39.2%
  • (on) Overcrowding, standing room only

    Votes: 100 18.9%
  • (never ever got on) Chinese escalator

    Votes: 221 41.9%

  • Total voters
    528
Status
Not open for further replies.
One thing we have to remember on performance is the following, according to nvidia Pascal took several billion to develop, amd would probably be lucky if a quarter of that was knocking around to develop their gpu's.

True but as above consumers don't really care about that but also even a tweaked and shrunk (albeit 28nm to 14nm FF isn't trivial to shrink) Fury would have been pretty competitive at this point giving them a bit more time to work on R&D for future generations. I really don't understand the reluctance at AMD to go with the leaner, faster architecture atleast on alternate cycles even if they want to push another approach in the long term.
 
True but as above consumers don't really care about that but also even a tweaked and shrunk (albeit 28nm to 14nm FF isn't trivial to shrink) Fury would have been pretty competitive at this point giving them a bit more time to work on R&D for future generations.

Especially with their Infinity Fabric they could have done something ala threadripper with those small Fiji dies. Unless they couldn't put infinity fabric on the Fiji die for some reason.
 
Especially with their Infinity Fabric they could have done something ala threadripper with those small Fiji dies. Unless they couldn't put infinity fabric on the Fiji die for some reason.

They wouldn't have been able to do anything different to current multi GPU - IF doesn't have the bandwidth or latency to tie monolithic cores together (plus all the latency at each end within the cores) as if one GPU for gaming purposes (though it should work great for compute workloads) and neither Fiji or Vega have the architecture to work in a "headless" fashion.

Looks like recent advances in substrate technology however is making it possible to produce a GPU in a manner of essentially laying out a monolithic core in a distributed fashion using separate packages on an interposer using a bunch of essentially "headless" GPU modules which going by recent news nVidia are looking at after Volta and AMD might be working towards similar capabilities as part of Navi's development. AFAIK though this requires far more single purpose dedicated interconnects than supported by IF which is a more general purpose widely flexible approach though it might be possible to implement it with some kind of dual mode functionality that lets it switch between those approaches as required.
 
Last edited:
you do realise nvidia/amd don't earn anymore from the inflated prices right? they sell at a stock price, it's the vendors that can charge whatever they want.

even if say, OCUK charged 2000 for 1080tis and they sold, Nvidia doesn't get a larger cut of profit, they get the same as if it was sold for £500
Im well aware of that. My point is if they sell out on everything they make. They are making good money.
 
On some FreeSync monitors the max refresh can only be attained with an adaptive sync capable output device which is something to look out for. You can mitigate some of the effects of not having G-Sync or FreeSync using nVidia's FastSync but its not a 100% plug and play solution - might find some incompatibilities or side effects with it in some games or just not working with some games.

I've just started using Fastsync and have a question, What am I meant to do regarding V-sync in the ingame menu? On or Off?
 
Fury all over again with pointless and expensive HBM that will limit availability, not that there will be a rush to buy em. AMD don't take the discreet gpu market seriously.
 
Fury all over again with pointless and expensive HBM that will limit availability, not that there will be a rush to buy em. AMD don't take the discreet gpu market seriously.
Yup. Volta or bust for me now. Maybe pick up a PS4 Pro in the meantime, if they drop the price below £300.

AMD just aren't at the races at all.

Who really wants a 375 watt (!!!!!!) 1080? I like my PC to be quiet. Fat chance if it's burning through that much power.
 
I've just started using Fastsync and have a question, What am I meant to do regarding V-sync in the ingame menu? On or Off?

While it should be over-rode by the nVidia drivers anyhow the best thing to do is turn off any V-Sync option ingame - you want the game to be able to run as fast as possible providing new frames as quickly as possible as the driver then selects the one that is closest to the V-Sync window to display reducing latency.

Somewhat counter-intuitively higher levels of max pre-rendered frames can actually produce better results here.
 
Last edited:
I know it's not quite the same, but just remember how much AMD have managed to completely undercut Intel on the cpu side. Here's hoping it's somehow possible in the gpu section too.

*edit* It's also possibly worth remembering that the 970 > 1070 price jump was over £100 at launch, the 980 >1080 price jump was over £150 and the 980ti > 1080ti was about £200. There may be some milking involved in those prices :p.

I think vega will be priced much lower than 1080 ... they need to factor in the 1080-like performance at a rumored 375W power draw.
Either ways if it has 15-18 billion transistors it is expected to age well.. just like the FuryX which had more transistors than its peers

The way things are going now..we may be looking at an uninspiring volta launch..nvidia will surely cut the GV100 in half and peddle that as enthusiast gear :)

It'd be nice if AMD did take the same approach they took with Intel but I can't see it happening, It'd be great but I think AMD are gonna have the attitude that this is a 1080ti competitor, even if it loses out more often than not. They priced the FE like a Titan and last gen they priced the Fury X like a 980ti even though it lost to it in just about every game benchmark on release, If you remember AMD came up with there own set of game test results on the eve of the Fury releasing that showed it beating the 980ti consistently, A couple of day's later when we had the independent test results we found out that AMD's results were false.
 
Especially with their Infinity Fabric they could have done something ala threadripper with those small Fiji dies. Unless they couldn't put infinity fabric on the Fiji die for some reason.

In all likelihood vega was already far past the point of production where it could have benefited from anything like that.
 
So now people are expecting Vega to be faster than the 1080 but at a cost of increased thermals and power draw.

If it's cheaper by at least £50 that's not bad. Hopefully more.

£100 would be about right for a product equivalent but a year later.
 
This just isn't adding up. Fury X Release Date: June 24, 2015

More than 2 years later they are releasing a HIGH-END card that is barely faster than the Fury X? Something just isn't adding up here.

Maybe I'm just in denial.
 
Im well aware of that. My point is if they sell out on everything they make. They are making good money.

They should be, It's selling faster than they can make replacements.


So as each day goes by the reviews etc are more and more damning.

This cant be right.


Have any reviewers done this type of review with a 1080ti and 980ti?
 
Status
Not open for further replies.
Back
Top Bottom