• 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 Radeon R9 290X with Hawaii GPU pictured, has 512-bit 4GB Memory

I really need a new GPU, my current one is a huge bottleneck... And now Nvidia announced the 780Ti which means I have to wait even longer.

BF4 with a 560Ti for almost a month won't be very fun :(:(:(
 
A 290X @ 1100mhz gets 15081 in 3DMark11, my 780 @ 1300mhz gets 14840. So the 290X is very fast. Not a worthy upgrade if you're on a 780, but for those looking now, it's a beast of a card. I wander how much faster the 780ti is going to be. Hope we get actual reviews of both these cards soon..

2prGSlU.jpg
PECwYfd.png

http://www.3dmark.com/3dm11/7346558
 
Last edited:
A 290X gets 15081 in 3DMark11, my 780 @ 1300mhz gets 14840. So the 290X is very fast. Not a worthy upgrade if you're on a 780, but for those looking now, it's a beast of a card. I wander how much faster the 780ti is going to be. Hope we get actual reviews of both these cards soon..

2prGSlU.jpg
PECwYfd.png

http://www.3dmark.com/3dm11/7346558

Very impressive considering the clocks on your card. With voltage control these cards are gonna fly. The gtx780ti is gonna have to be a monster to better that. It's probably gonna be a case of nvidia using higher clocks to be faster at stock but most likely parity when both are clocked. A titan running at 1100 core would probably not match that.
 
Last edited:
I badly want a r9 290X, but I'm holding out for the non reference cards to appear, especially if the card makers increase the memory, etc, before I make a decision.
 
Very impressive considering the clocks on your card.

Yup, looks like the 290x is very fast, only a couple hundred points in between cards here, but how much further can the 290X go?

Sidegrade for 780 owners, but at least the pressure is back on Nvidia, they will want to retain the crown, so hopefully the 'ti' will be a full 2880 cores. Roll on reviews..
 
Of course if Mantle does take off then it could see the 290X pull ahead of the 780/TI/Titan in games, even if benchmarks don't reflect it.

Of course by the time Mantle has really caught on the 290X may well be a generation or 2 out of date...
 
Of course if Mantle does take off then it could see the 290X pull ahead of the 780/TI/Titan in games, even if benchmarks don't reflect it.

Of course by the time Mantle has really caught on the 290X may well be a generation or 2 out of date...

The more interesting thing about mantle for me is the potential for the last gen cards like the 7970/50 to achieve gtx780/titan performance if the gains are big enough. If the gains are big enough and enough games use it nvidia may finally have to start competing on price as atm there brand is strong enough to let them price there cards way above amd.
 
The more interesting thing about mantle for me is the potential for the last gen cards like the 7970/50 to achieve gtx780/titan performance if the gains are big enough. If the gains are big enough and enough games use it nvidia may finally have to start competing on price as atm there brand is strong enough to let them price there cards way above amd.

Mantle isn't a straight up performance accelerator - you can't just plug it into a game and immediately see a say 20% boost in performance.

Most games use a variety of techniques for working around draw call limitations - running these techniques on top of mantle would probably result in a small performance increase but it would be mostly lower single digit percentages. These techniques are fairly fundemental to most engine designs so I can't see anyone considering it worth the work rebuilding existing engines to use mantle instead and even if they did you'd end up more or less back at the same level of performance just brute force rendering now instead of using an elegant work around.

In future engines it could have more impact especially as it would allow you to turn effects upto 11 before seeing the old performance limitations - obviously mantle is more than just about draw calls, it allows you to tinker with things at a lower level making effects possible that couldn't be done or done in realtime before - but as to some degree that potentially locks features out from non-AMD hardware I can't see the potential there being fully realised.
 
Why don't we just wait for December when Mantle get stitched into BF4 before we say "its this and that and it will do this and that..." no one knows any of that, no point in pretending we do.

OcUK must be getting their stock this week, unless they already have it. Gibbo?

Waiting for the ;) thread.

No point in asking Gibbo, he will just go "NDA on the NDA for the NDA"

If they are not released in time to be in a customers hands before BF4 AMD will get murdered.
 
Last edited:
I think they will be since BF4 is out in the US on the 29th, and people have ordered the Battlefield 4 edition :P

Although there's nothing to stop them releasing it after the game is out is there?
Where the Asus Rampage IV Battlefield 3 editions out on BF3 release day? (Honest question, I've forgotten)

I know it's not quite the same, but was every game in the AMD Never Settle pack released on the same day AMD made it available in the bundle?
 
Sure hope we at least get a release date this week,

My first foray into 1440p starts tomorrow, my intention was to get two 280x's and use them.

But can't help the feeling that I'm buying 'old' tech, and with the New leaked benchmarks I'm thinking I should have just waited a little longer and gone for a 290/290x.

Ill see what they perform like and dsr them if I'm not happy.
 
Mantle isn't a straight up performance accelerator - you can't just plug it into a game and immediately see a say 20% boost in performance.

Most games use a variety of techniques for working around draw call limitations - running these techniques on top of mantle would probably result in a small performance increase but it would be mostly lower single digit percentages. These techniques are fairly fundemental to most engine designs so I can't see anyone considering it worth the work rebuilding existing engines to use mantle instead and even if they did you'd end up more or less back at the same level of performance just brute force rendering now instead of using an elegant work around.

In future engines it could have more impact especially as it would allow you to turn effects upto 11 before seeing the old performance limitations - obviously mantle is more than just about draw calls, it allows you to tinker with things at a lower level making effects possible that couldn't be done or done in realtime before - but as to some degree that potentially locks features out from non-AMD hardware I can't see the potential there being fully realised.

So basically you think if a game is coded on mantle you will be able to make a game look better and still have the performance compared to what is possible on dx11. The only reason i think mantle is going to give a performance boost in games was the statement from amd saying a r9 290x is going to make titan look slow in comparison in bf4. If this is the case the 7970 is surely going to be close to top end nvidia cards like the gtx780.

I was at my mates last night and his boy put on a few old school games on the ps3 namely gta 4 and infamous. It totally shocked me how bad they looked in comparison to gta 5. If mantle can extract this kind of enhancement then it definitely has some good potential. i know it's not all down to the api but it's amazing what the programmers have been able to extract from this very dated tech.
 
Last edited:
Back
Top Bottom