ASUS R9 280X DirectCU II TOP 3 GB

[W1zzard]

ASUS is using a completely revamped PCB design and the cooler from their GeForce GTX 780 DC II on their latest R9 280X card. The card also comes overclocked out of the box with an overclock on both memory and GPU, which provides a serious performance advantage over the reference design.

Show full review

 

[dj-electric]

This. This is how you make an R9 280X. Thanks.

 

[1d10t]

7970 GE very well re-branded by Asus

 

[..0]

the fan noise testing is faulty, it freely adjustable. how about finding the lowest fan rpm to hit 79c then also publish that information, that would actually been something....

 

[Casecutter]

Well you can buy this and a nice 2650x monitor all while staying within spitting distance of a GTX 780, it's crazy! There's no reasoning to a GTX780 pricing (or a R9 290X if priced that high) for single standard monitor gaming. Then the GTX770 is no deal, needs huge price adjustment. W1zz gave most GTX770 9.5-9.8, which brought nothing more than this is doing and held a much higher pricing (less reduction). Remember you're getting for $310 is 3Gb of 384-bit, which is very much a boon and it shows.

The one thing I notice is looking at the Asus GTX 770 DirectCU II OC 2 GB W1zz did a while back, Asus or Nvidia does seem to have a much better control for cooling at idle, in both temp and noise. However at load the R9 280X for 1dbA more it cools some 6°C lower, all the while dissipating 11% more power. The weird part is the R9 280X cooler has 5 heat pipes, while the GTX 770 DirectCU has 3 heat pipes.

The R9 has what appears to use exactly like the Asus GTX 780 DirectCU II OC 3 GB with that "new" Asus CoolTech dual fan blade, then the second old traditional fans. How does that setup work so well on the 780 at 28dbA/35°C and then 32dbA/32°C for the R9? It seems weird that Asus could make the 780 as quiet and cool, while R9 isn't. I suppose it boils down to the desktop clock profiles (135 vs. 300Mhz), but the Asus 780 is 15W idle verses the R9 at 12W (25% higher). While load is also strange Asus 780 is 210W/67°C/30dbA (new bios), while the R9 is 208W/70°C/35dbA... Something doesn't make sense the fan runs faster though doesn't cool as well for the same or less heat load?

I think the biggest problem that AMD did... Was not doing something with the R9 280X in least how to employ a improve PowerTune profile. While better the 7970Ghz AMD needed to tune it further, as Nvidia did with GPU Boost 2.0 and desktop clock profiles.

 

[Footman]

As always a nice review. Personally I am holding out for a nice new R9 290X.....
BTW looking at the breakdown of the HSF I noticed that Asus put far too much TIM on this GPU and that only 3 of the 5 heatpipes make contact with the gpu.......

 

[Casecutter]

BTW looking at the breakdown of the HSF I noticed that Asus put far too much TIM on this GPU and that only 3 of the 5 heatpipes make contact with the gpu.......

Noticed the same thing on the ASUS GTX780 DirectCu II OC, figure either has area approaching 1/5 the die on the 10mm H-P perhaps a little better for the 780, but yes neither is optimal or touch the two outboard tubes. Asus design the cooler for both, you can see the tapped holes on the 780 picture that the R9 uses and visa-versa.

http://www.techpowerup.com/reviews/ASUS/R9_280X_Direct_Cu_II_TOP/images/cooler2.jpg
http://www.techpowerup.com/reviews/ASUS/GeForce_GTX_780_Direct_Cu_II_OC/images/cooler2.jpg

Hoping to see the Gigabyte version (their GTX780 was 25dba idle), plus normally has thermal pads for the memory.

 

[W1zzard]

Gigabyte version

waiting on an updated bios from them to fix fan issues

 

[Olleing]

A reflection about. Asus card! Bios switch is missing, in other words. there is only a bios

The absence has not been featured in a few other reviewers.

http://www.techpowerup.com/reviewdb/...s/AMD/R9-280X

 

[The Von Matrices]

W1zzard, you should really modify the wording in all your R9 and R7 reviews where you state:

You may use all outputs at the same time, so triple-monitor surround gaming is possible with one card.

This is misleading. The 280, 270, 260, (and presumably 250) are the same dies as their predecessors and have three TMDS links but still two TMDS clock generators. AMD has only modified their VBIOS so that two TMDS displays with the same timings can share the same clock generator. Even then, AMD software limits this capability to only work when all three TMDS outputs have the same clock signal.

This means that you can only use all the outputs when you have three identical TMDS monitors.

NVidia's solution is much more flexible since it has three clock generators and three TMDS links, so all the displays can be mismatched without issues. The current phrasing implies that AMD has similar display output capability as NVidia, which is not true. With mismatched monitors, AMD cards can still only use two TMDS outputs. I suspect it may cause a few headaches for people who buy these cards expecting all the outputs to work with mixed monitors when in reality they won't. Maybe the R9 290 series will fix this problem.

 

[sweet]

This card makes the 9.6-9.7 scores on the rebrand 770 look like a joke

 

[1d10t]

W1zzard, you should really modify the wording in all your R9 and R7 reviews where you state:
This is misleading. The 280, 270, 260, (and presumably 250) are the same dies as their predecessors and have three TMDS links but still two TMDS clock generators. AMD has only modified their VBIOS so that two TMDS displays with the same timings can share the same clock generator. Even then, AMD software limits this capability to only work when all three TMDS outputs have the same clock signal.

This means that you can only use all the outputs when you have three identical TMDS monitors.

NVidia's solution is much more flexible since it has three clock generators and three TMDS links, so all the displays can be mismatched without issues. The current phrasing implies that AMD has similar display output capability as NVidia, which is not true. With mismatched monitors, AMD cards can still only use two TMDS outputs. I suspect it may cause a few headaches for people who buy these cards expecting all the outputs to work with mixed monitors when in reality they won't. Maybe the R9 290 series will fix this problem.

This unfold AMD had difficulties syncing all output since the HD 6k days.They offer two DVI-D,one HDMI and two mini DP,down to single DVI-D,one HDMI and two mini DP in 7k era and now they opt for two DVI-SL,one HDMI and one DP.
I suspect AMD is playing exonerative by 'limiting" display capabilities,and focusing on frame pacing issues by adopting PCIe link bus.

 

[PatoRodrigues]

this card makes the 9.6-9.7 scores on the rebrand 770 look like a joke

+1

 

[manofthem]

Thanks W1zz

 

[jormungand]

great review of this card, msi failed in his gaming edition but asus did a nice job here. great card for $310.

 

[AlderaaN]

waiting on an updated bios from them to fix fan issues

Hello W1zzard, and thank you for the review.

Been a long time reader of techPowerUp, glad to be here.

I've got a few burning questions if you please, regarding what's quoted above:
My new gaming rig (listed in my specs) is missing only its GPUs.

I'd love getting myself two GIGABYTE R9 280X GV-R928XOC-3GDs,
but after seeing these fan noise results - I'm quite reluctant in doing so just yet.

1. Have you been in contact with GIGABYTE about this and can confirm they're actually aware of the issue?
2. If so, will such a BIOS be made available to the public any time soon?
3. Will you also be providing a review of said GPU in the near future?

Many thanks for your time!


Regards,
//Subscribed to thread.

 

[W1zzard]

Hello W1zzard, and thank you for the review.

Been a long time reader of techPowerUp, glad to be here.

I've got a few burning questions if you please, regarding what's quoted above:
My new gaming rig (listed in my specs) is missing only its GPUs.

I'd love getting myself two GIGABYTE R9 280X GV-R928XOC-3GDs,
but after seeing these fan noise results - I'm quite reluctant in doing so just yet.

1. Have you been in contact with GIGABYTE about this and can confirm they're actually aware of the issue?
2. If so, will such a BIOS be made available to the public any time soon?
3. Will you also be providing a review of said GPU in the near future?

Many thanks for your time!


Regards,
//Subscribed to thread.

not exactly the right thread, but ok I'm testing the card right now and I've been talking to gigabyte, they have already sent me a new bios, which reduces fan speed, from too noisy to still too noisy. they explain this, with the Asian market that focuses on temperatures more - they say ..

I will post any BIOS from any mfgr that I'm told to use and publish results with (won't post a BIOS that they send me to just test and feedback)

 

[AlderaaN]

not exactly the right thread, but ok I'm testing the card right now and I've been talking to gigabyte, they have already sent me a new bios, which reduces fan speed, from too noisy to still too noisy. they explain this, with the Asian market that focuses on temperatures more - they say ..

I will post any BIOS from any mfgr that I'm told to use and publish results with (won't post a BIOS that they send me to just test and feedback)

Sweet. Looking forward to the review.

I'm guessing there won't be much chance in seeing another BIOS from there that's even less noisy..


P.S.
Sorry about the wrong thread thing, thanks for going with it anyways

 

[MrYellow]

Hey, good read. Thank you.

Are the fans audible @ idle? Even if pushed in fan control profile to minimum?

 

[Ravenas]

Requesting a review for the following card:

SAPPHIRE 100363TXSR Radeon R9 280X 3GB 384-bit GDDR5 PCI Express 3.0 x16 HDCP Ready CrossFireX Support Video Card

 

[AlderaaN]

Requesting a review for the following card:

SAPPHIRE 100363TXSR Radeon R9 280X 3GB 384-bit GDDR5 PCI Express 3.0 x16 HDCP Ready CrossFireX Support Video Card

Dunno if that's the thread to ask for it, but that would be sweet!

 

[Ravenas]

Dunno if that's the thread to ask for it, but that would be sweet!

Given that the ASUS is considered to be his top pick, I would love to see how it stacks up versus the Sapphire variant which most reviews have considered to be better. The thing I like about W1z's reviews is that they are very detailed. I don't always agree with him, but they are still very detailed. Just a request.

 

[Mattriz]

Given that the ASUS is considered to be his top pick, I would love to see how it stacks up versus the Sapphire variant which most reviews have considered to be better. The thing I like about W1z's reviews is that they are very detailed. I don't always agree with him, but they are still very detailed. Just a request.

as you can see here and here the sapphire version has better frames in games, but the asus version uses less power, has less noise and lower temperatures.

 

[Steevo]

Why does it show at PCIe 1.1 instead of 3.0? Power management, a glitch, or actually not running at full PCIe speed?

 

[Ravenas]

as you can see here and here the sapphire version has better frames in games, but the asus version uses less power, has less noise and lower temperatures.

I would like to see a review with W1z's level of detail. That's basically what I was saying.