• 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.
I just can't comprehend how it 'should' have been so much better based purely on the clockspeed than Fury X, but somehow isn't. WTF AMD ?
 
JediFragger said:
Let's wait til the proper launch with stock/benchies tbh. They're probably whipping their poor driver guy to death making him work 24/7 on them . Poor ******* :D
Click to expand...

I’m about 99% sure this is the reason for the delay. RTG driver team are having to rewrite the driver for TBR and it’s taking longer than expected.

So not exactly sandbagging as such, just a case of the card needing a few more weeks to ripen.

Can I just add to the posts above -

WAIT FOR BENCHMARKS

(there are a lot of you speculating but the fact is, the driver guys have two more weeks to sort this out)
 
CAT-THE-FIFTH said:
Did you see that slide - it has better support for mining. The Vega 56 looks to have a significantly lower TDP than the 64 and is not massively slower in pure FLOPs either. You know what is going to happen if it is good at mining.
Click to expand...

Oh crap, means the cards will get snapped up by miners.

Seems GLoFo's 14nm really doesn't like high clocks without killing power consumption either.
 
Tute said:
I’m about 99% sure this is the reason for the delay. RTG driver team are having to rewrite the driver for TBR and it’s taking longer than expected.

So not exactly sandbagging as such, just a case of the card needing a few more weeks to ripen.
Click to expand...

They wouldn't have done this event comparing any performance if that was the case.
 
AMDPower said:
its a shame really, as a AMD fan I am seriously disappointed. Many of us kept telling AMD what they needed to achieve to make this acceptable to us gamers and it would seem AMD dont understand how much of a fail this card will be. They really needed the XT version to be 10% faster than 1080 and XTX 20% faster then 1080, they could have justified the prices they are charging. Nobody in their right mind would pay these prices specially given the hideous power draw unless they are stuck with freesync monitor.

I really hope they have some magic drivers up their sleeve somewhere but that's not happening. They should have not released this card and worked on getting Navi done faster.
Click to expand...



I'm sure AMD hoped they would get the performance figures as you suggested but their engineering or the process failed for some reason.

Anandtech said that according to AMD most of the additional 3.9bn transistors in Vega over Fiji was designed to increase the clock speed by increasing pipeline stages etc. That isn't a bad deign choice in itself but it does mean you are royally screwed if you can't hit those high clock speeds. This is like what Intel did with the P4, they had a design that would massively increase clock speed and they had exceptions of hitting 5Ghz within a couple of years. Of course they couldn't hit those clock speeds so the whole architecture was let down.

Vega seems to draw way too much power, and this is what is preventing the clock speeds. Maybe AMD had much higher exceptions of the efficiency improvements that their simulation models were indicating. Maybe the Global Foundaries process really is liming them.


Either-way, i'm pretty sure AMD didn't envision this kind of performance. A die that big,, expensive GBM2 memory and interposer, 15 months late and there is just no reason for Veg to exist really.
 
Scougar said:
I just can't comprehend how it 'should' have been so much better based purely on the clockspeed than Fury X, but somehow isn't. WTF AMD ?
Click to expand...

If you read the AT article on Vega they recently did AMD used nearly 3.9 billion transistors to be able to clock it sufficiently high, and that essentially the increase in transistors from Fiji to Vega there,so it makes me wonder whether having less memory bandwidth than a Fury X has really not helped.
 
humbug said:
I have a Free-Sync screen, i have had this dilemma a while ago and got sick of waiting on AMD, so i ended up with one of these. its so much faster than the 290 its beyond funny...

Here's one for sub £400 https://www.overclockers.co.uk/pali...ddr5-pci-express-graphics-card-gx-03n-pl.html you will not regret it.
Click to expand...

Isn't that 480 quid :p?

I don't really want to pay that money for that performance now. I could have done that last year.
Thats the dilemma, anything I buy now will last less time than it would have done had I bought last year.

Plus I don't want to buy Nvidia really :p. I'm happy to spite myself, but I'm not going to pay amd the same money for the same performance either
 
Scougar said:
I just can't comprehend how it 'should' have been so much better based purely on the clockspeed than Fury X, but somehow isn't. WTF AMD ?
Click to expand...

They lowered IPC, and added more transistors; all to boost clocks as much as possible. The die shrink helped with that and lowering power consumption.

It just shows what dire straights AMD was in after they nearly scrapped their discrete GPU section to fully focus on consoles and APUs.

I really hope they can sort Navi out before it's too late. Raja only joined, and was put in charge on RTG in late 2015, after Fiji was launched and Polaris done. Let's see what his team can really do from scratch; as back in his "glory days" he was there with ATi everything from the 9700Pro to 5870 before moving to Apple.
 
CAT-THE-FIFTH said:
If you read the AT article on Vega they recently did AMD used nearly 3.9 billion transistors to be able to clock it sufficiently high, and that essentially the increase in transistors from Fiji to Vega there,so it makes me wonder whether having less memory bandwidth than a Fury X has really not helped.
Click to expand...

That bandwidth really intrigues me as well, as the reduction from 4 to 2 chips is what someone else mentioned as well when I was asking why the bandwidth was so low.

I actually wonder if they originally thought they would have 4 chips, but decided to reduce to 2 due to costs or availability of HBM2, and then screwed themselves for performance? Overclocking the memory on Vega should help see if that may be the case.
 
CAT-THE-FIFTH said:
If you read the AT article on Vega they recently did AMD used nearly 3.9 billion transistors to be able to clock it sufficiently high, and that essentially the increase in transistors from Fiji to Vega there,so it makes me wonder whether having less memory bandwidth than a Fury X has really not helped.
Click to expand...
It probably hasn't helped, bu AMD coudln't go with a 4 stack design again. HBM just isn't ready for GPUs yet.

The 1080ti offers 480GB/sec as it is.
 
Well in my case Pascal Titan with block should be here this week. Thanks AMD for making it easy for me :)
 
Status
Not open for further replies.
Back
Top Bottom