Anandtech 290x review, 780 killer, Titan's Rival, for $549

[GoodBytes]

a 780 is 670 bucks where i am and the r9 290x is 549. Assuming you already have a custom loop it comes out to the same price because a ekwb for it is 100 dollars. so you get titan performance for 650 bucks. 

 

Considering that the great majority of people have air cooled, you have to assume that one have nothing related to water cooling, and that is my point. Now if you have watercooling already.. then sure! That will be a good buy.

[leipero]

Not until AMD do something about the cooling.  The card is being throttled badly when it gets warm, so Unless it can be better cooled then there is nothing that software, drivers or mantle will do to improve the figures.

Custom cooling solution, problem solved, on the side note, even with throtling 290x is quite a bit faster. Also, AMD could do something about cooler design etc. but that would bring price up, then you would complain about proce, i think this is better trade off.

 

True PcPer showed that the normal avergae clock is around 820mhz and not 1ghz as AMD says.

http://www.youtube.com/watch?v=e6zzvUihPbw

1Ghz is not advised frequency that "AMD says", they said "Up-to 1GHz", that means, well, up to 1GHz, and that statement is true.

[mr moose]

Custom cooling solution, problem solved,

SO to the cost of the card you add a $100 water block, another $300 for radiators, fans, reservoir and pump. Not to mention the controller and the experience needed to set it up and you have spent more than titan and we don't even know if aftermarket cooling is actually going to allow it to perform better.

 

No thanks.

[leipero]

SO to the cost of the card you add a $100 water block, another $300 for radiators, fans, reservoir and pump. Not to mention the controller and the experience needed to set it up and you have spent more than titan and we don't even know if aftermarket cooling is actually going to allow it to perform better.

 

No thanks.

No need for water block, MSI and ASUS (for example) custom cooling versions will have, well, custom cooling solution, it is nothing new to the market, in last 10 years there was custom PCBs and cooling solution, and reference solutions was always as bad as it can be (in both AMD and nVidia history)..

[mr moose]

No need for water block, MSI and ASUS (for example) custom cooling versions will have, well, custom cooling solution, it is nothing new to the market, in last 10 years there was custom PCBs and cooling solution, and reference solutions was always as bad as it can be (in both AMD and nVidia history)..

no non-reference designs for the foreseeable future, I am hoping this will change, but not likely.  We will probably need to wait until either a WC overclocker does some bench testing or until asus, et al get their act together and get some stuff out to the reviewers. Which I've heard rumours it will be quite sometime away.

 

http://www.techpowerup.com/191704/radeon-r9-290x-launch-date-revealed.html

[Godlygamer23]

Last I check it's a 2 stage system.

For true near 0 watts, the display needs to be off.

Isn't that what I just said? I said "completely." Which means it is working, just not all the way, until the display is off.

[KylSp]

 

Yup. This is the most concerning part. I can understand a card that goes to 94C in FurMark, but 94C in gaming?! I would be sweating if I saw those temps. The price and performance we see in the benchmarks is great though. I'm sure non-reference coolers will keep this card to more reasonable levels.

 

 

 

Do you realize how many people are running Titans and 780s as is? It's completely realistic to expect people to run the card as they got it from the box. Otherwise they should just sell it as a bare PCB.

 

I would buy all of my cards bare PCB if that was actually an option. The cost of the cooler on the reference titan / 690 / 780's is anything but cheap, and having assembled a few systems with those cards I can say that material wise it absolutely increased the cards manufacturing price. 

 

It all depends on what segment of the market you are in for what product points you care about. People went nuts about the 480's with reference coolers, but on water they were a completely different story. I hope the same thing applies to the 290x, that when not restrained by the relative mediocrity of modern video card air cooling deltas, we'll see what it can really do. That doesn't abolish what the out of the box performance characteristics are, but again, not everyone will be running them stock.

 

If you aren't going water, wait for the non-reference models.

 

 

** Some of you can't understand the concept of an enthusiast, that if someone is spending $600 on a card, another $100 for a waterblock to integrate into an existing loop probably doesn't matter at the end of the day. This is a major hobby for some people, and just because you scoff at what a water cooling system costs doesn't make that opinion meaningless.

[DLM_012]

This is a beast card performance wise but as others have said heat is a serious issue. 94C is insane. 

 

I'm also concerned with noise. I'm waiting for the day when graphics cards are as silent as Noctua fans (sub 25Db) while still cooling them within reason. Then and only then will I die happy.

[JAKEBAB]

Fml anyway mantle isn't going to be a big thing lol I don't know why everything still thinks it will be

[KylSp]

Fml anyway mantle isn't going to be a big thing lol I don't know why everything still thinks it will be

At best it will achieve glide-level adoption, which on one hand is good as it may push MS to actually improve and not version lock directx, but it creates market divides between camps which is not healthy for the consumer. At worst it flops and EA wasted a few months of development time adapting Frostbite3 for mantle. 

[afa7336]

I said that you COULD buy a waterblock, and that you should buy a 3rd party card instead, because buying the reference card would be stupid. 

 

Okay, can you send me a link to a 3rd party card I can buy right now? There is none as far as I can find, until there is, you can't say go buy something that doesn't exist. I bought reference cards before, does that make me stupid? Nope. I threw in Reference cards into a friends system before, I guess that was so stupid of me, since it was a single card and she isn't a hardcore gamer. Damn it, I guess I should apologize to her for being stupid right?

 

And you do know that while we here love 3rd party cards with custom cooling solutions (i.e. DirectCU II, etc.), not everyone buys them. You have system builders that use reference design, you have people that get reference because they are either cheaper or they don't know better, etc. The market is a weird place, and if you expand upon what you know, you'll find out that this forum and its users is actually a niche in the market.

[mr moose]

 

, but again, not everyone will be running them stock.

 

That's the problem, right now we don't have an option, they can only be run at stock.

 

 

If you aren't going water, wait for the non-reference models.

 

If the rumours are true, by the time they hit the market Nvidia will have another option out.  So AMD might miss the boat.

 

 

** Some of you can't understand the concept of an enthusiast, that if someone is spending $600 on a card, another $100 for a waterblock to integrate into an existing loop probably doesn't matter at the end of the day. This is a major hobby for some people, and just because you scoff at what a water cooling system costs doesn't make that opinion meaningless.

 

The number of enthusiasts running a custom water loops is minimal,  And even if it wasn't, we are still talking about adding a whole new layer of cooling gear to gain an unknown performance boost when there is already a card out there with proven performance on it's reference cooler.

[Matman24]

If I had the need and the disposable income for it, I'd get the 290X, but I'd wait for the aftermarket coolers to come out (probably a Windforce or a Twin Frozr card). 

[t0ps]

Some valid points being made here, I keep imagining the performance underwater, but that price/performance factor disappears. Hopefully the aftermarket coolers will do something about that heat so we can get a proper comparison going.

[AraSarkisyan]

Fml anyway mantle isn't going to be a big thing lol I don't know why everything still thinks it will be

it really is.......

[AraSarkisyan]

One thing is a bit odd. People are telling how fast the 290X is at 4K resolution but does somebody actually have a 4K screen? It seems a bit silly that people are so happy about the relatively low price tag but then the display costs similar to really high end gaming rig. So basically very few of us will ever use 290X with 4K display. By the time 4K displays are anywhere near affordable, 290X is going to be wastly outdated.

 

And the other matter is comparing 290X to Titan? WHY? I've seen many people comparing 290X and Titan together especially on gaming performance. Why would you compare Titan to 290X when Titan is not really any faster in games than 780. Even many reviewers say that if you bought a Titan for gaming, then you basically wasted money.

its wonderful for multiple displays

[LAwLz]

it really is.......

Can you please tell me next week's lottery numbers before you put your magical crystal ball away?

No but seriously, nobody can know how good Mantle will be. Personally I have been too disappointed too many times to blindly believe the hype and will remain on the fence until I see reliable third party numbers.

[afa7336]

it really is.......

 

So, where have you played with Mantle, I wonder? Can you explain your first hand experience with Mantle? How much of an FPS gain did you get from it, did Nvidia cards perform just as well without Mantle, did you see an impact on any other parts of your system from using Mantle, how did it compare in crossfire versus SLI? 

 

I'm not holding my breath on an answer from you, mind you, but considering there has been NO release yet with Mantle integrated, we don't know jack squat about Mantle, so no one including you, can say what the outcome of Mantle will be.

[JAKEBAB]

it really is.......

Lol if u actually think mantle is going to be big u need to look at the facts developers arnt going to develope for another API also they have no reason to as next gen consoles won't be supporting mantle! Microsoft has already said they will be "tweaking" direct x which already has a massive adoption and a lot easier to code for plus it will have the advantage over being a fairly low level API if Microsoft sticks to what they said.

[beebskadoo]

Here's my opinion. The reason AMD hasn't released any performance numbers on mantle is because it's not the game changer people think it's going to be. So why not hype the shit out of it before the newest gen launch to boost sales? This could very well be what Is going on. Might not be but just my thoughts.

Sent from my SGH-T999 using Tapatalk

[ViruZ_HUN]

OHH YES YES: MY HOPES IS BACK!: THX:D

 

not i know r9 290x beats 780:D

[That Norwegian Guy]

LTT is dissolving under the sheer weight of these kiddie posts... Almost not worth it anymore

[beebskadoo]

LTT is dissolving under the sheer weight of these kiddie posts... Almost not worth it anymore

Which ones ?

[ACatWithThumbs]

LTT is dissolving under the sheer weight of these kiddie posts... Almost not worth it anymore

It's so sad he just said how it is.

The reference design is not worth to buy.

Who wants a GPU that 95c and is so loud that you can't hear your game anymore.

And he is right you can already get a GTX780 that is quiet and cool for 13€ more.(And with overclock it's faster)

EDIT:

I was really hoping this card to be a TITAN killer so Nvidia has to get Maxwell out as soon as possible with a low price.

But now Nvidia has nothing to worry and can just make the prices crazy high again like with the 700seires

 

[xAcid9]

Just slap a Prolimatech Pro MK-26 if u guys can't wait for non-ref. It only cost under $60.