[HardOCP] Asus DC II 290X max overclock versus GTX 780ti max overclocking review:

Page 9 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BallaTheFeared

Diamond Member
Nov 15, 2010
8,115
0
71
I think its odd people actually believe its the cooler causing the sample varience.

I also had to laugh at the "beast" Tri-X that was linked to on OCN running 1.4v, 100% fan in a valley run. He'll be lucky to get 1150 fully stable with 1.35v and a lot of noise.
 
Feb 19, 2009
10,457
10
76
Nothing odd at all given the take down notice from Gigabyte admitting a batch was faulty.

Also nothing odd at all looking at the actual heatpipe contact on the die. I think GK110 would be just as "varied" if 2 of its heatpipes are broken.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
For some reason I think inadequate cooling would lead to throttling. No? Is that the logical conclusion? Just thinking out loud here. There are two water cooling users at H who are using 1.4V and can't get past 1100mhz. What it seems like to me is that AMD pushed the chip hard out of the gate to compete with nvidia and it doesn't have a lot of headroom to go past that. Some might get lucky. But overclocks with the Hawaii are all over the place and not consistent, even with over-voltage.

Oh, and, the card that H tested didn't throttle. Despite having 1.35V pushed through it. If the cooler was truly faulty the card would throttle or crash. But that didn't happen.
 
Last edited:

omeds

Senior member
Dec 14, 2011
646
13
81
What are you saying Silverforce? Outside of a single sku there are no 290x's that run properly? That doesn't bode well, if 99% of models don't run optimally.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Nothing odd at all given the take down notice from Gigabyte admitting a batch was faulty.

Also nothing odd at all looking at the actual heatpipe contact on the die. I think GK110 would be just as "varied" if 2 of its heatpipes are broken.

If a cooler is inadequate and can't cool a chip within specifications variance doesn't happen. Throttling happens. BSODs happen. TDRs happen.

None of these things happened with the H test despite 1.35V being pushed through it. You also keep throwing out mentions of the Tri-X card. Does sapphire have a major presence in AU? It's not like any result with the tri-x overclocked would be different, it would still lose vs an overclocked 780ti, like you already stated.
 
Last edited:
Feb 19, 2009
10,457
10
76
What are you saying Silverforce? Outside of a single sku there are no 290x's that run properly? That doesn't bode well, if 99% of models don't run optimally.

What I am saying is of the custom boards so far, ASUS and Gigabyte are failures, due to improper usage of heatpipe coolers not designed to fit Hawaii.

That leaves only Sapphire Tri-X, XFX DD and MSI Gaming. These are the ones currently available.

Soon there's PCS+, HIS.
 
Feb 19, 2009
10,457
10
76
If a cooler is inadequate and can't cool a chip within specifications variance doesn't happen. Throttling happens. BSODs happen. TDRs happen.

None of these things happened with the H test despite 1.35V being pushed through it. You also keep throwing out mentions of the Tri-X card. Does sapphire have a major presence in AU? It's not like any result with the tri-x overclocked would be different, it would still lose vs an overclocked 780ti, like you already stated.

So why is it [H] with its 500W and 90C temps didn't scale in Crysis 3 when other review sites have shown scaling for R290/X in that game. It also happened for FC3 for [H]. Two out 4 games, its broken. It certainly looks crap if you read [H] and assume it must be normal.. but alas, other sites show the scaling is fine.

So, as a gamer, do you bother to question why is it [H] and Computerbase.de had an ASUS R290 that had issues with overheating or requiring high fan speed and still running hot... or you just ignore that and go ahead and think its fine.

Remove yourself from the bias for a second.

Take a look when a review does their OC vs OC with cards that aren't overheating.
http://www.guru3d.com/articles_pages/msi_geforce_gtx_780_ti_gaming_review,27.html
GTX780ti 1228mhz boost. Great scaling, great performance.

The same ASUS R290 but apparently this one work fine, 50% fan speed for a huge OC without 90C temps.
http://www.guru3d.com/articles_pages/asus_radeon_r9_290x_directcuii_oc_review,28.html
Again, great scaling, great performance.

Yes, the 780ti OC is faster. We already knew this. What did you hope to accomplish exactly when its already known the 780ti GPU is the fastest since its launch.
 

omeds

Senior member
Dec 14, 2011
646
13
81
What I am saying is of the custom boards so far, ASUS and Gigabyte are failures, due to improper usage of heatpipe coolers not designed to fit Hawaii.

That leaves only Sapphire Tri-X, XFX DD and MSI Gaming. These are the ones currently available.

Soon there's PCS+, HIS.

Oh right, there's 3 sku's, making 90% of the available 290x's duds? :hmm:

Interesting, since even reference 780/ti seem to be ok.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
I'm not even sure what you're going on about here SF. Just answer these questions. Does the Hawaii throttle when cooling is inadequate? Yes or no? If VRM temps go dangerously high, willt he Hawaii throttle or BSOD/crash? Yes or no?

Your implication was the cooler was inadequate. IMO, if it was cooling with 1.35V being pushed through the chip it was doing just fine. But please. Yes or no answers to the above questions. And then moving on, when we realize that H didn't get throttling, what does that imply?

Either you can't accept the fact that Hawaii isn't overclocking too well or AMD released faulty chips out on the market that aren't throttling as intended. But let me know if Hawaii is supposed to throttle when cooling is not adequate. A simple yes or no answer will suffice. If the answer is no, I guess some news outlet will have to do investigative reporting on whether AMD is releasing faulty chips on the market which aren't throttling as they should.

Oh yeah. As far as the computerbase test. Looks like SKYMTL of hardwarecanucks has that answer covered:

http://www.hardwarecanucks.com/foru...ii-oc-review-comment-thread-3.html#post749649
 
Last edited:
Feb 19, 2009
10,457
10
76
For my own cards, yes and yes, to clockspeed throttling. I am not there to see how [H] does it or how well they monitor it.

But if they get 90C with a weak OC at 75% fanspeed when other sites get much less temps at 50% fanspeed, it tells me they have a dud. Just like another reputable review site, Computerbase.de, both their ASUS and Gigabyte were duds and surprise, shock and horror, Gigabyte admitted it was their fault. Asus is keeping it mums, but looking at their heatpipes not making contact.. its no surprise at all.

Are YOU denying that having a heatpipe cooler with 2 of its pipes not making contact is bad for its performance?
 

Spidre

Member
Nov 6, 2013
146
0
0
33 min 137 max for tomb raider.

Come on tressfx...

Also will be interesting to see lightning vs lightning.
 
Feb 19, 2009
10,457
10
76
So AMD released faulty chips on the market that don't throttle as intended. That's your answer?

For my own cards, yes and yes, to clockspeed throttling. I am not there to see how [H] does it or how well they monitor it.

..................

Are YOU denying that having a heatpipe cooler with 2 of its pipes not making contact is bad for its performance?
 
Feb 19, 2009
10,457
10
76
He's just pointing out that even if no pipes touched it wouldn't matter performance wise unless the GPU throttled, which it didn't.

I would certainly love to see their clockspeed graph because it makes no sense to have next to no scaling when other review sites clearly show it has no issues in those games.

Even the crap reference card, when its not throttling scales with OC.
http://www.techpowerup.com/reviews/AMD/R9_290/29.html
 

el etro

Golden Member
Jul 21, 2013
1,581
14
81
So AMD released faulty chips on the market that don't throttle as intended. That's your answer?

No. Probable problems on 290 performances are BIOS and/or powertune and PCB/energy system of the card. Hawaii chip never was/is causing any of 290 cards problems.
 

BallaTheFeared

Diamond Member
Nov 15, 2010
8,115
0
71
Sad when AMD non ref boards are completely outclassed by nvidia reference boards that people called cheap, and lacking power delivery.
 

skipsneeky2

Diamond Member
May 21, 2011
5,035
1
71
Sad when AMD non ref boards are completely outclassed by nvidia reference boards that people called cheap, and lacking power delivery.

Can't wait to see what the MSI 290x Lightning does when it debutes, have always had a thing for those and the Matrix but never gave in to buying one.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Here's the conclusion since some people here CLEARLY didn't read the article. Here is Brent's conclusion. I suggest reading it entirely. It covers all the incorrect statements that were mentioned by several posters in this thread.

We have learned a lot with our time spent overclocking and gaming on the ASUS R9 290X DirectCU II OC and the GeForce GTX 780 Ti. Our overclocks may seem lower than other sites, but that is because we will not accept any room for instability or artifacts after prolonged sessions of gaming. We want to achieve an overclock that a gamer would set, and forget. We don't just run benchmarks, or timedemos to stress test, we actually play the game, whole levels, whole chapters, even half the game itself at times, to find out what is stable. Multiple hour long sessions of Battlefield 4 Multiplayer are common practice for testing overclocking.

We have definitely determined that "warming up" every video card has to be done to properly get a handle on a video cards real in-game performance for prolonged gaming sessions. This goes for the AMD R9 290X and the GeForce GTX 780 Ti as well, which does have a dynamic clock.

We found that the GeForce GTX 780 Ti would start off at 1006MHz real-time frequency in games for about ten minutes, then slowly drop to a level that it mostly stayed at the rest of the time. If someone were running a benchmark, or timedemo, which usually exist inside that timeframe, it would be showing inflated performance. It is not until you let the cards heat up and game with these for at least 30 minutes that clock speeds settle down.

We found this an important part of testing and finding the best overclock for our video card. What might be stable for ten minutes of gaming, wasn't for hour long sessions of gaming.

The ASUS R9 290X DirectCU II OC video card never throttled on us while overclocking, unless we turned down the power target or fan speed too much. As long as we kept the fan high enough to achieve lower than 94c temperatures, the GPU would not throttle. Our overclock has the temperature at 90c, which was perfect, no throttling at all.

Overclocking Results
For the ASUS R9 290X DirectCU II OC we managed to overclock the video card to 1115MHz with a 1.35v setting and 5.67GHz memory. Remember, the video card is already overclocked at 1050MHz versus 1000MHz on a reference card.

The reference cooler and reference cards are notorious for throttling performance. We experience no throttling what-so-ever on the ASUS R9 290X DC2 OC video card while overclocking. This means we saw a consistent 1115MHz clock speed in every game.


It is important though that the temperature remain below 94c in order to achieve said non-throttling. That means a fairly high fan speed of 75%, which is not a leaf blower, but certainly on the level of, if not slightly louder than the reference cooler fan in Uber mode. The GeForce GTX 780 Ti fan doesn't reach that level till around 80%, but any higher is definitely not going to be liked by most anyone.

Basically, 100% fan on both cards is unusable in a real-world situation. No one would game like that. Unfortunately for the R9 290X, it really needs better cooling to raise the voltage even higher.

Overclocking on Air Has Its Limits with R9 290X
We had our ASUS R9 290X DC2 OC running at 1.35v, and it took 75% fan just to keep it from going over 90c. That's a lot of fan just to keep it from the throttle point. The R9 290X GPU is not built for extreme overclocking unless you are on water, or have some super duper custom air cooled card that isn't out yet. On water we could see 1.4v being a reality, and perhaps that will allow you to edge a few more MHz out of it. On air, 1.35v should be your limit, I'd be worried about damaging the GPU at anything above that on air.

Low MHz Mega Hurts
While our overclock is low compared to previous generations, we think it has more to do with the GPU's limits than the video cards. The video card seemed right capable of pushing the GPU harder. However, the GPU just wouldn't take it.

In our Performance Variance article we noticed how even among the exact same cards hardware differences can create different performance profiles. If the AMD R9 290X is that sensitive to performance differences imagine how much variance there must be in each different R9 290X GPU itself. You may get one that just won't' budge much on overclocking, and from the same manufacturer you might get another one that overclocks by leaps and bounds.

This is what we think is going on, variance among GPUs. If you thought overclocking varied per GPU in the past, we think the condition is even more prevalent on the AMD R9 290X. Hawaii just wasn't meant for 28nm, the architecture inside is ahead of the packaging process available. If only 22nm were ready we think the R9 290X would be showing itself in a whole different light.

The Bottom Line
The bottom line is this, on our first attempt at overclocking a custom AMD R9 290X based video card, we are left wanting. We get the impression that the AMD R9 290X GPU is just not good at being overclocked. We base this on the fact that it consumes an abhorrent amount of power as you increase voltage, which in the process creates a crap ton of heat that needs to be displaced. All of this has to be done just to achieve a decent overclock out of it.

The fact that the AMD R9 290X video card needed a custom card configuration and custom cooling solution just to run at its stock "up to" default 1GHz without throttling is not a good sign. That then limits the potential for overclocking the GPU since the custom card and cooler are already working overtime just to run the card at its normal clock speed.


It will take some massively grotesque custom cooling on air to get the most potential out of the GPU from overclocking and also not sounding like a jet taking off. There are many manufacturers who do partake in the ultimate enthusiast based graphics card, ASUS being one of these with its MATRIX line. We don't' know if that is planned, but it is going to take that level of video card and cooling to get the most out of the AMD R9 290X on air. What an absolute beast of a GPU, and we don't mean that in a good way.

None of this is ASUS' fault, ASUS has designed a stellar video card with the ASUS R9 290X DirectCU II OC. We love this video card from an engineering standpoint in every which way. We are literally drooling at the lip over the amount of design and high-end components that went into this video card to give the AMD R9 290X the best chance it can have in a two slot design.

The only issue with the ASUS R9 290X DC2 OC is the MSRP $569.99 price tag, and the unavailability of it still. (Editor’s Note: As we are going to publish, we have finally seen a price of $699 from Newegg.) It is due out this month, but currently prices are jacked up beyond reason due to high demand of R9 290/X for mining. Hopefully these prices will stabilize and fall back down to where they need to be, but in the mean time, it does mean a more expensive purchase than it was intended to be.

If pure performance is your thing, it looks like an overclocked GeForce GTX 780 Ti may be in your future, which is showing to sell for $719.99 with Free Prime Shipping. We have a retail card undergoing a full evaluation, and that should show us the potential of what a custom card can provide. If overclocking is not your thing, the ASUS R9 290X DirectCU II OC still competes quite well with the GTX 780 Ti, beating it most of the time. Just don't overclock the 780 Ti and embarrass the AMD R9 290X GPU, that way your R9 290X GPU will stay all warm and fuzzy inside.
 
Last edited:

Skurge

Diamond Member
Aug 17, 2009
5,195
1
71
Of course now [H] overclocks are the only ones that matter.

Still I'm not sure what the point of this thread is. We all know the 780Ti is the faster card in stock form and when overclocked. Although overclocked performance didn't matter when GTX680 came out. Whatever though.

You still have to admit that the issue with Hawaii is keeping it cool and being able to feed it enough power.

As a side note. The Reference boards overclock just as well as this custom Asus card. I also saw that [H] overclocks for GTX780s were in the same range so yeah it guess they are still pretty close.
 
Last edited:
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |