• 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 Working on At Least Three Radeon RX Vega SKUs, Slowest Faster than GTX 1070

1gxCHM3.gif
 
They are always behind and they always will be. For us AMD guys that are invested in Freesync and been waiting since 290/390/480 type performance, this will still be a good leap. I don't need it to batter a 1080ti, I just need it to be a nice leap over what I've got, which even if it's 1080 speed then it will be. Will be perfect that for me on a 1440p freesync screen, if that price is around 400 so what if it's a year late, we AMD guys haven't had that option for years. Yes I know we had a Furyx but don't even get me started on that POS.

I was doing the same think waiting and then decided to get g sync monitor and nvidia card. Yes of course I wanted to pay less but how long can you wait.
 
Last edited:
i think with HBM2, AMD might have around 20% less power draw, that should offset the deficit they have against pascal, if Vega is to be any close to 1080Ti perf, it would come at the same power draw as an OC Ti, AMD tend to push the clocks as much as they can.
 
Last edited:
i think with HBM2, AMD might have around 20% less power draw, that should offset the deficit they have against pascal, if Vega is to be any close to 1080Ti perf, it would come at the same power draw as an OC Ti, AMD tend to push the clocks as much as they can.

But don't forget, it'll only offset GloFo's poor(er) manufacturing techniques (vs TSMC/Samsung) a little.
 
20% is also high, id say 10-12% (30w) and Pascal is far more effecient than that relative to Polaris. Polaris is not much better than Maxwell despite being on a smaller node. Pascal has twice the performance per watt as the RX580. If AmD gain a 50% efficiency improvement which would be huge give the same process node, they would still behind even with HBM.
 
I was doing the same think waiting and then decided to get g sync monitor and nvidia card. Yes of course I wanted to pay less but how long can you wait.

Thankfully some of us have more will power. I've recently bought a UW 1440 freesync monitor as I refuse to pay Nvidia prices for g-sync, so now I'm waiting on a AMD GPU capable of powering it. So I'll wait as long as it needs
 
Thankfully some of us have more will power. I've recently bought a UW 1440 freesync monitor as I refuse to pay Nvidia prices for g-sync, so now I'm waiting on a AMD GPU capable of powering it. So I'll wait as long as it needs
In my case it was 8 years since i upgraded, but i manged to scoop g sync monitor for cheap, and gtx 1070 miss-priced so it costed me around £650
 
No it's not but Nvidia i preparing Volta which will be even quicker than 10 series. It looks like AMD is always behind.

They have been for a while due to AMD struggling to keep it's head above water,
The worst is over, or at least it should be and now they should be in a position to
spend more on gpu research meaning RTG should get more competitive
over the next few years.

There's a lot of shoulds in that paragraph but I've got my fingers crossed for the 'shoulds' to
become wills, are's & have's.
 
That's pretty much the levels of performance they need if they want to compete considering they're launching these a long time after 1070/1080 launch.
 
I was doing the same think waiting and then decided to get g sync monitor and nvidia card. Yes of course I wanted to pay less but how long can you wait.

That's the biggest problem but it's only one if you're desperate to spend money on overpriced hardware,
Yes I want an upgrade and yes it feels like it's taking forever to get here but my current hardware is
perfectly capable of playing games still. And that makes me think
"So what if I have to wait twice as long for AMD upgrades"
If like me someone hasn't got money to burn AMD is the better option, Even though they're going
to make us pay as much as Nvidia do for the highend card ie: big Vega - 1080ti
If it repeats the sequence of the last 2 years it will remain sat at the top of AMD's pile and relevant
in AMD's priority structure for longer than cards currently do in Nvidia's,
Especially important if you goal is to buy a high end card and keep it until it's no longer able
to run games well enough,
A good example is the 290x and 780ti which came out within a few weeks of each other.
New games today bring the 780ti to it's knees and sometimes even the 1050ti is ahead of it
where as the 290x still does a respectable job.
Even with all these factors taken into account there's no guarantee that AMD will continue
to work the same way as they have for the last few years.
The tables may turn and Nvidia may look like the saint for a while but in the here and now
AMD is the more consumer friendly option so I'm happy to remain with them and their slower
release system. That said once Nvidia have good api support and Freesync support I'll happily
jump onto one of there cards,

My guess is this:
Once Nvidia's dominating market share is in decline and they want to sell a new range of cards
that range will offer Freesync support while older ranges won't, At least not until they've
persuaded the masses to replace their Pascal and Volta cards with cards from the new
Freesync capable range.

It may sound like I'm being negative to Nvidia but I'm not it's just business and that's what
they are good at.

Thankfully some of us have more will power. I've recently bought a UW 1440 freesync monitor as I refuse to pay Nvidia prices for g-sync, so now I'm waiting on a AMD GPU capable of powering it. So I'll wait as long as it needs

Ditto,
Once Nvidia support Freesync I'll happily go green again but for now I'm AMD only.
 
20% is also high, id say 10-12% (30w) and Pascal is far more effecient than that relative to Polaris. Polaris is not much better than Maxwell despite being on a smaller node. Pascal has twice the performance per watt as the RX580. If AmD gain a 50% efficiency improvement which would be huge give the same process node, they would still behind even with HBM.

glofo is pretty efficient as long as you dont push the clock way beyond a certain point, where power draw become crazy compared to clock and performance scaling, RX480 at ~150watt, and GTX 1060 at ~ 120watt, that's nowhere near twice the efficiency.
as long as AMD doesn't give in to the pressure and OC Vega beyond what the node allow it to, efficiency should be within 20-30watt from 1080/ti equivalent, 220-250watt for nvidia to 250-300watt for AMD, that's what i am hoping for at least, then if ppl want performance over efficiency they can OC, but AMD shouldn't make the choice for them.
 
My guess is this:

Once Nvidia's dominating market share is in decline and they want to sell a new range of cards that range will offer Freesync support while older ranges won't, At least not until they've persuaded the masses to replace their Pascal and Volta cards with cards from the new Freesync capable range.

That would be smart, but would they then ditch Gsync altogether? Probably not, as it would upset the installed user base that paid dearly for their monitors...

Does Gsync require any special hardware on the GPU side? Or is that special chip in monitors all that is needed and the GPU just needs to send some signalling via software/bios?
 
glofo is pretty efficient as long as you dont push the clock way beyond a certain point, where power draw become crazy compared to clock and performance scaling, RX480 at ~150watt, and GTX 1060 at ~ 120watt, that's nowhere near twice the efficiency.
as long as AMD doesn't give in to the pressure and OC Vega beyond what the node allow it to, efficiency should be within 20-30watt from 1080/ti equivalent, 220-250watt for nvidia to 250-300watt for AMD, that's what i am hoping for at least, then if ppl want performance over efficiency they can OC, but AMD shouldn't make the choice for them.

But then surely they are making the efficiency over performance choice for them then?
 
Back
Top Bottom