Is custom air cooling enough for the R9 290X?

[Orangey]

I've not heard anyone defending the DCu II so I assume Asus bodged it and did something stupid like re-used the Nvidia cooler on the smaller Hawaii core.

Wait for a rev2 or however many revs it takes for a proper fix then compare. Of course by then the Lightning et al will be out.

 

[uksoldierboy]

How'd you do that? I just cannot get my R290's to SLi with my GTX780

(I know you didn't mean it like that )

Lol black magic sonic

 

[Gregster]

So people would rather buy a Asus DCII and ignore this and the [H] thread and anything else that basically shows this card up?

1115mhz at 1.4v is shocking and I know full well that other cards are capable of more mhz on less volts. I would be glad of this thread if I was looking to buy this card but seems some don't want people to know and would rather call me out.... Seriously?

I didn't realise I needed to butter it up for the sensitive men here

 

[pgi947]

You also have to consider this may make other customers and reviewers take a closer look at other custom cooled solutions for the 290 series, for all we know other cards may be suffering from the same fate. Its certainly clear that at least one board partner has taken the heat output of these chips a little for granted.

 

[Gregster]

You also have to consider this may make other customers and reviewers take a closer look at other custom cooled solutions for the 290 series, for all we know other cards may be suffering from the same fate. Its certainly clear that at least one board partner has taken the heat output of these chips a little for granted.

There is a review of the Sapphire Tri-X on Eteknix that paints a similar story. They only managed 1110Mhz on their overclock but it isn't as in depth as the [H] review.

http://www.eteknix.com/sapphire-r9-290-tri-x-4gb-graphics-card-review/18/

It would be good to know what volts they used and what temps were. The card looks great in fairness but if that is the standard or just another poor example is any one's guess.

Edit:

Techpowerup paint a similar picture for the DCII and got even less on the overclock with only 1090Mhz on their maximum overclock.
http://www.techpowerup.com/reviews/ASUS/R9_290X_Direct_Cu_II_OC/27.html

Powercolour 290X reached a clock of 1100Mhz, with the top card tested being the AMD reference card with clocks of 1145Mhz.

 

[LtMatt]

11% overclock on stock voltage sounds about right. I can achieve a similar overclock on my cards. Are you on a smear campaign Gregorio? The Sapphire Tri X are excellent cards that are cool and overclock well, commented by people using them here and in reviews. The voltages used below is not a lot at all. Tiny amount of voltage used by Anand.

By using TRIXX we were able to get the Sapphire R9 290 Tri-X OC we were able to hit 1251MHz on the core and 5666MHz on the GDDR5 with .085V of extra voltage. This vastly outstrips the levels attained by our reference samples and brings actual performance to some incredible levels.

Source
http://www.hardwarecanucks.com/foru...72-sapphire-r9-290-4gb-tri-x-oc-review-8.html

Starting first with gaming performance, as a more extensive overclock relative to Sapphire’s factory overclock, the performance gains from our own overclocking have yielded very solid results, despite the fact that this isn’t explicitly an overclocking board. Between the 13% core overclock and 15% memory overclock, the average performance increase measures in at 12%, varying depending on whether a game is more bandwidth limited, GPU limited, or CPU limited.

Source
http://www.anandtech.com/show/7601/sapphire-radeon-r9-290-review-our-first-custom-cooled-290/5

 

[Gregster]

11% overclock on stock voltage sounds about right. I can achieve a similar overclock on my cards. Are you on a smear campaign Gregorio? The Sapphire Tri X are excellent cards that are cool and overclock well, commented by people using them here and in reviews. The voltages used below is not a lot at all. Tiny amount of voltage used by Anand.



Source
http://www.hardwarecanucks.com/foru...72-sapphire-r9-290-4gb-tri-x-oc-review-8.html



Source
http://www.anandtech.com/show/7601/sapphire-radeon-r9-290-review-our-first-custom-cooled-290/5

From your same link.

For overclocking the 290 Tri-X OC, we increased the PowerTune limit by 20% and the voltage by 50mV to what’s roughly 1.23v. Out of this we were able to get another 125MHz (13%) out of the GPU and 800MHz (15%) out of the VRAM, now topping out at 1125MHz for the GPU boost clock, and 6GHz for the VRAM. The final clockspeeds are better than our previous efforts at overclocking our reference 290 (which was prior to voltage control), although only moderately so.

So they got to 1125Mhz at Anandtech, which is good I assume?

 

[LtMatt]

From your same link.



So they got to 1125Mhz at Anandtech, which is good I assume?

Yes, because they got a nice overclock out of the memory for little increase in voltage. Anandtech increased voltage to 1.23v, the limit is 1.4v. I think you should probably look to own the cards in question Greg before trying to label them as a poor example or of a low standard.

 

[LtMatt]

Dunno what Gregsters obession with trying to prove AMD cards are so bad is.. get out more man!

Yet another great thread in the GPU forum.

Tbh i'm tired of seeing the same crap by people who don't even own the cards.

Same can be said for those slating nvidia who've never owned a nvidia card in their life. People saying PhysX is crap who've never even used it, commenting on nvidia drivers when they've never installed them, and to think someone told me the other week that this section has calmed down, lol.

This thread is obvious baiting the title says it all by not including the specific ASUS model.

I have to agree with all of the above.

 

[Rangerjr1]

I get a feeling that most reviewers dont know what they're doing when they have an AMD card. They have throttling issues because its configured that way, if you dont want it to throttle just let the fan speed increase...

 

[Rangerjr1]

From my own experience Nvidia cards seem to be a better purchase if you wanna plug and play. You gotta work with the AMD cards to make them shine, i think i have proven this in benchmarking. But i guess thats what you get when you pay less for a card.

 

[Gregster]

Yes, because they got a nice overclock out of the memory for little increase in voltage. Anandtech increased voltage to 1.23v, the limit is 1.4v. I think you should probably look to own the cards in question Greg before trying to label them as a poor example or of a low standard.

I didn't do any labelling, that was all done for me but it is your opinion and you are welcome to it. The reviews I linked show that 1145Mhz is the maximum clocks they reached and if this is good, I don't see the problem. You showed how the DCII doesn't have adequate cooling, so surely this is part of the reason the 290X gets so hot when trying for big clocks? The way I see it is avoid this card.... Look to the Lightning if you want to give it some with the 2 8 pin connectors and higher quality components and possibly binned chips. At £530 for the DCII, I would hope for something that is a bit special and this is clearly not the case from the [H] review anyways.

From my own experience Nvidia cards seem to be a better purchase if you wanna plug and play. You gotta work with the AMD cards to make them shine, i think i have proven this in benchmarking. But i guess thats what you get when you pay less for a card.

It does seem that way. Maybe the reviewers were not using the tools properly or they did just have rubbish clockers, as looking about, many users are hitting 1200Mhz in bench tests.

 

[Rangerjr1]

Oh and btw, new 290x incoming. Got my RMA approved . Should be interesting Gregster.

 

[darket]

From my own experience Nvidia cards seem to be a better purchase if you wanna plug and play. You gotta work with the AMD cards to make them shine, i think i have proven this in benchmarking. But i guess thats what you get when you pay less for a card.

+1 , I'm in no way an nvidia fanboy, however i have had no regrets returning my 290x and replacing with a 780ti, its just a much better card.

As i've said before the architecture feels as if its at its limits, and in the silicon lotery there are more misses than hits.

The 290 on the other hand is just about right, i could never reccomend a 290x over a 290.

 

[Gregster]

Oh and btw, new 290x incoming. Got my RMA approved . Should be interesting Gregster.

What make of 290X was it? I am still waiting to hear back from Asus about my Titan that was sent on the 12th of December

Ohhh and I expect you will be giving it some?

 

[Rangerjr1]

+1 , I'm in no way an nvidia fanboy, however i have had no regrets returning my 290x and replacing with a 780ti, its just a much better card.

As i've said before the architecture feels as if its at its limits, and in the silicon lotery there are more misses than hits.

The 290 on the other hand is just about right, i could never reccomend a 290x over a 290.

Is your 780ti faster than my 290?

 

[RavenXXX2]

I get a feeling that most reviewers dont know what they're doing when they have an AMD card. They have throttling issues because its configured that way, if you dont want it to throttle just let the fan speed increase...

They review the cards as is that goes for Nvidia and AMD, it's not up to reviewers to do AMD's work for them.

 

[Rangerjr1]

What make of 290X was it? I am still waiting to hear back from Asus about my Titan that was sent on the 12th of December

Ohhh and I expect you will be giving it some?

Its the Sapphire version... Sadly.

I hope this one has Hynix, and yes ill whip it some

 

[Rangerjr1]

They review the cards as is that goes for Nvidia and AMD, it's not up to reviewers to do AMD's work.

Its reflected in its price i feel. People who know what they're doing get most out of AMD cards, thus not being the optimal plug and play cards.

 

[Steampunk]

I don't even see the point of comparing overclocking between brands. Surely if a card can't overclock much even if it's staying within it's thermal limits, then you're getting nearly the max possible out of the card, which is good, right? Who wants to buy a card that leaves tons of performance on the table for which you have to mess with voltages and careful tuning?

Comparing overclocking ability is even worse than comparing clockspeeds between brands. It doesn't tell you how much work each card is doing per clock, so a slower clocked card could have higher performance.

How is any of this even useful unless your only aim is to see how far above stock speeds you can get the card, ignoring if one is already most of the way there?