• 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.
Vega will have probably taken a backseat to Ryzen, that'll have been their main focus, their CPUs, so hopefully, now that those are sorted, we might now get the GPUs sorted, and see that happening after Vega.

opethdisciple said:
You are right. Nvidia build products for today and AMD build for tomorrow. It's always been like that. I think it comes down to budget. AMD have to hedge their bets wisely so bet on tech that will be valuable tomorrow.

For example bulldozer. At the time the market needed IPC not cores! So it was poor. BUT.... it paved the way for Ryzen.

Nvidia have huge RnD budgets. No doubt Volta will be a beast. But Vega.... AMD don't have the resources to do now and tomorrow so from what I see have thought let's to a card for tomorrow at least it will stretch out for some time.

Is the market ready for Vegas features? Nope. No games need HBCC, FP16 etc.... Heck not many games actually use DX12 or Vulkan properly and yet Vega is much better in these regards than the 1000 series.

I'm sure Volta will fix a lot if these issues. But I imagine as HBCC is hardware if they don't already plan to release a gaming cards with such tech it won't be coming in the next series. (If it even takes off)
Click to expand...

Ive always said they should stop the new tomorrow, and just concentate on the here and now like Nvidia, as they need market share, only doing for the now will get them that, then once they have it, they can then start to push the new tommorow.

We've seen it with everything they do, it just gets no up take, because they have little to no share, devs are just not interested in something, that only a few people will be able to do, they just won't waste their time.
 
Last edited:
B12B6 said:
Well that is not what he said. And if you want to buy Volta or need more performance than what Vega can give ofcourse then there is no point. But how many people need that? 10? 1000? A lot of people are gaming old slower cards so why not buy freesync? Like I said already, its very few people who need that cind of performance. Not everyone can drop 500£ on gfx card. So when they buy card + monitor, that gsync/freesync difference might be what matters. Not to mention in tech there is always tommorrow when new and better is right around the corner. If you want to wait, you can, if you dont, you buy whats sold today.
Click to expand...


Well not many need more power. I'm one of them my 980ti barely keeps wow over 30fps in 3D and its just enough for my VR

and for ,If you want to wait, bit kinda we waited ******* 14 months already !!!
 
I'm almost trying to convince my self that Vega is the way to go. :p I haven't made any decisions yet until I see reviews. But... I go and spend £500 on a 1080... Volta will be hear soon with the features that Vega has. :o
 
On top of that you've got this controversy that Nvidia drivers are no good scaling across multi cores.

Which I think deserves another deep dive. Because if it's true TRUE that's a big deal. Again something I expect them to fix as of Volta.

Reiterating my thoughts that Nvidia build for today and AMD build for tomorrow.

When multi cores are wide spread and the average Joe has 6+ cores in his rig Nvidia will reengineer their drivers to scale better.

AMD knew with Ryzen and with Bulldozer that the average Joe would soon have affordable multicore chips in their rigs so put the effort in too DX12/Vulkan and scaling side ways rather than IPC.
 
melmac said:
That's not really true though is it? You have a Freesync monitor now and you need 1080 performance, you buy a Vega card to go with your Freesync monitor. It's going to be a hell of a lot more expensive to sell the freesync monitor and buy a gsync monitor. You are also going to be paying the higher prices from Nvidia too for those 5 years.

Volta could be another 6/7 months away. And if it follows the same release pattern as Pascal, then the Volta X80 cards will be at Ti prices for 6 months. So you could be waiting a year more to buy.
Click to expand...
I agree in that you've stated the case for existing free sync owners but to actually go out and buy one now and be stuck with vega performance would be madness imo.
 
My issue with current 1000 series is I was suspicious that they didn't have the DX12 performance. Which is why I didn't jump on the bandwagon. Every game performs better in DX11 than DX12 with Nvidia.
 
opethdisciple said:
On top of that you've got this controversy that Nvidia drivers are no good scaling across multi cores.

Which I think deserves another deep dive. Because if it's true TRUE that's a big deal. Again something I expect them to fix as of Volta.

Reiterating my thoughts that Nvidia build for today and AMD build for tomorrow.

When multi cores are wide spread and the average Joe has 6+ cores in his rig Nvidia will reengineer their drivers to scale better.

AMD knew with Ryzen and with Bulldozer that the average Joe would soon have affordable multicore chips in their rigs so put the effort in too DX12/Vulkan and scaling side ways rather than IPC.
Click to expand...
If I had a Ryzen PC I would just go for Vega at this point, then can always upgrade to Navi a couple years later. At 1440p Vega would be more than enough I would imagine. If I was on 1440p I would be getting a Vega 56, but Vega does not look like it will cut the mustard at 4K and feels like it would be a waste of my money when Volta is around the corner which will likely provide 1080Ti performance for around Vega 56 prices and run a hell of a lot more efficiently making less noise. So I will miss out on Freesync, so what, never bothered me all these years, won't let it bother me now :p
 
Just a thought, with Pascal being just a clock bumped Maxwell, as many AMD supporters have told us and Maxwell being released way before Ryzen was, how is it NVidias fault that their architecture doesn't play well with Ryzen's multiplier core architecture. Why isn't it AMD's fault for building a CPU that deliberately hampers Nvidia's cards.

Well we have to argue about something while we wait another 2 weeks till we finally see just how much of a Dick Fosbury Vega is. :D:p:D
 
opethdisciple said:
AMD have always been first to market with features:

1. 2Gb Vram. My 5870 had it, which is why I bought it over the GTX 470.
2. Eyefinity, first iteration of proper multi monitor support.
3. DX 10.1 again AMD first.
4. Vulkan in the guise of Mantle developed by AMD.
5. DX12 feature sets more support by AMD than Nvidia.
6. Now we have HBCC and FP16 hardware support.
Click to expand...

Being picky the first 2GB VRAM cards were nVidia a number of 3rd party made 2GB 200 series cards, etc. but it would have been a bigger advantage on the cards contemporary with the higher end 5000 cards - I'll concede AMD were first to utilise it effectively on GPUs.

nVidia had multi-monitor gaming support (span view or whatever it was called) then they pulled it as a professional card only feature - Eyefinity forced them to put it back on GeForce.

DX10.1 is a mixed one - like many things they try to do they tried to push 10.1 before GPU performance was good enough to be worth supporting it though on the flip side nVidia DX9 wasn't upto scratch and like with later Mantle stuff (which pushed them into producing their enhanced DX11 driver) it forced nVidia to double down on their efforts.

I guess you could say there are many things we can be thankful to AMD for even if not quite the way you meant it :D
 
opethdisciple said:
My issue with current 1000 series is I was suspicious that they didn't have the DX12 performance. Which is why I didn't jump on the bandwagon. Every game performs better in DX11 than DX12 with Nvidia.
Click to expand...
Just as well given there are only 3 dx12 (only) games!

Look Polaris banked on dx12 and it didn't happen.

Vega may bank on dx12 but it's not going to happen in its 'prime'. Maybe five years down the line it will have a few more fps than a 1080 in dx12 but at which point nobody will care.
 
Too me it feels like the 1080 is a card for today's games and Vega a card for tomorrow.

But you know what I think they do this on PURPOSE.

Buy a card for today's games and tomorrow you can buy another one for tomorrows games etc etc etc....

It's done on purpose!
 
Brazo said:
Just as well given there are only 3 dx12 (only) games!

Look Polaris banked on dx12 and it didn't happen.

Vega may bank on dx12 but it's not going to happen in its 'prime'. Maybe five years down the line it will have a few more fps than a 1080 in dx12 but at which point nobody will care.
Click to expand...

You may have a point.
 
If we look at the next 4 big games (I'm interested in), I think it's safe to assume the scores are tied at 2-2.

Far Cry 5 and Wolfenstein are AMD games and Metro and Assassins Creed will be Nvidia games.
 
opethdisciple said:
My issue with current 1000 series is I was suspicious that they didn't have the DX12 performance. Which is why I didn't jump on the bandwagon. Every game performs better in DX11 than DX12 with Nvidia.
Click to expand...

Current feature set:

yiN9JZg.jpg
 
Status
Not open for further replies.
Back
Top Bottom