Partner Cards: One Radeon R9 290 And Three 290Xes, Reviewed

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

IndustryStandard

Honorable
Nov 7, 2013
44
0
10,540
Yay! I've been waiting for a (any) partner R9 review for months. I'm going to loosen my tie and grab another cup of water before I waste office time to read this.
 

Gurg

Distinguished
Mar 13, 2013
515
61
19,070


Probably so much lower than the others that they were hidden by the trees. The next logical question is: how much additional performance can be generated if the reference 290x was overclocked to just the level of the lowest of the rest.
 

bemused_fred

Honorable
Feb 18, 2012
519
0
11,010


This whole mess could have been avoided if they'd released after-market cooled cards at launch, and so made them actually worth buying back then.
 
Good read, thanks for the hard work and time invested to produce it. For me, I'm still happy with my 3x 6950's and I havent even felt the need to switch the 2nd BIOS on to run 6970 mode. Looks like I'll sit this release out and see if the next batch of cards can convince me to upgrade.
 

9-Ball

Distinguished
Apr 20, 2010
111
0
18,690
.... and I grow happier and happier with my 2 780 Ti's. I've used ATi/AMD video cards since the 3800 was new but I'm happily skipping this latest 'generation'. Anyone weighing a 290x vs 780Ti purchase, as I did for most of November and December, I would heartily recommend the latter, especially the EVGA OC'd model.
 

eklipz330

Distinguished
Jul 7, 2008
3,034
19
20,795
so i guess the reference 290/290x plus watercooler/kraken is the most viable option, especially if it comes down to less than what sapphire is offering, am i correct?
 

ezareth

Distinguished
Dec 15, 2005
41
0
18,530
As an owner of an Asus R9 290X Directcu ii Card I'd like to say that Tom's either got a bad card for use in this test, or they did their testing after removing the heatsink and didn't reapply it correctly or something. Their temperatures are not consistent with what I'm seeing at all and if you look at other review sites Tom's temps are higher than those reviews as well.I looked closely at both the Sapphire Tri-X and Asus Directcu before deciding on the Directcu. I've played farcry 3 at 2560X1600 for 6 hours run utilizing ultra settings and I further overclocked the card to 1110Mhz and 5750 on the memory yet the card never crossed the 81C and usually sits at 80C (At 75% fan speed). When playing at stock settings the card is rock steady at 78C with stock fan speeds. I don't think the fact that all 5 heatpipes aren't touching the GPU is really relevant as it has full contact across the entire GPU. I agree the Tri-X cooler is superior to the Directcu, but that doesn't make the Directcu bad in any way.If you are considering an non-reference 290X there are some things about the Tri-X you should be aware of. For one it only comes with a 2 year warranty (While the Asus has a 3-year) and two, unlike what the article says it is a 3 slot card, not a 2 slot card. I'm planning on acquiring a second 290X for Crossfire once I find a suitable 4K monitor to game with and crossfire Tri-X due to the 3 slot form factor is not an optimal solution on most setups.Lastly on the noise, I can barely hear my card over the case fans on my Antec 1200 at 75% speed, when gaming I can't hear it at all. This card is far quieter at 75% speed than the 6970 Sapphire reference cards I replaced with it. Calling this card "Relatively Loud" does this site a disservice. It is possible either the case they were using had airflow problems (which would be surprising) or the card itself they tested on had some sort of problem (being at higher temperatures due to an issue would naturally ensure a higher fan speed)Just my two cents. Take this review with a grain of salt.
 

FormatC

Distinguished
Apr 4, 2011
981
1
18,990
The results were proofed by Asus - with the same conclusion. I'm disassembling and assembling cards since 18 years and I know what I do :)

Fact is (take a look at the heat sink) that the cooler is a failed construction. Sorry for this news, but it is a fact. If you have a case with a lot of fans inside, you can get lower temps - may be. And: the Tri-X is a clear 2-slot design, I've measured the card by myself.

radeon-r9-290x-cooling-case,X-W-415940-22.jpg


This is really idiotic. Please tell me, how many heat pipes have contact to the GPU.
 

ezareth

Distinguished
Dec 15, 2005
41
0
18,530
Ok I spent a little more time researching this (which as a reviewing site I would have expected you guys to do instead). The Sapphire Tri-X does not have a "Quiet" mode or performance mode switch on the card. Instead it is a UEFI Boot mode switch. However, you stated at the beginning of the review "All of the Radeon R9 290X cards are set to their Quiet Mode firmware setting" which in the case of the Tri-X does not exist. What this means however is that the Asus card in quiet mode is set to allow its temperatures to rise to 94C. By nature of doing a comparison at this point you should have the other cards to their performance settings which is equal to the default Tri-X setting.

Secondly, Sapphire lists the Tri-X on their site: as a 2.2X which is the same as a 3 slot card. You can't put them back to back (not that anyone would do so with a non-blower design).

http://www.sapphiretech.com/presentation/product/product_index.aspx?cid=1&gid=3&sgid=1227&pid=2090&lid=1

Most importantly to me your results do not match the temperatures I am seeing on my existing card, in my existing setup. I have an Antec 1200 which is full tower case, but the fans are all stock and the case is not known for exceptional cooling. I do not think the results of your review are representative of what someone buying this card would expect in a normal air-cooled case.

As far as your assessment of the heatsink, have you ever seen a CPU heatsink that exactly matches the size of the CPU? No, they're all much larger (including the heatsink on the Tri-X) and that doesn't make the heatsink bad, it indicates waste on the part of the design that is contributing to the cooling of the card less, but in this case more heatpipes than come in contact with the GPU doesn't translate to the design being bad. I'm not saying the Tri-X cooler isn't superior, I'm saying labeling the Directcu cooler as "Too Hot" when you have it set to a setting that is DESIGNED to operate hot and"Relatively loud" when it is far quieter than the reference design is not very objective.

Ultimately the Asus DirectCU is overclocked higher than the Sapphire Tri-X so it is going to run hotter. There is just too little information about the cards and their different settings given in this review to make it particularly useful.



 

rdc85

Honorable
Some manufacturers has giving a wc variant but it still need proper loop to utilize..

IMO this is better than they give AIO closed loop WC.. since it provide more flexibility, and more cooling capacity...
 

jenniferrhicks

Honorable
Feb 3, 2014
7
0
10,510
m­­­­­­­­­­­­­­­­­­y­­­­­­­­­­­­­­­­­­ n­­­­­­­­­­­­­­­­­­e­­­­­­­­­­­­­­­­­­i­­­­­­­­­­­­­­­­­­g­­­­­­­­­­­­­­­­­­h­­­­­­­­­­­­­­­­­­b­­­­­­­­­­­­­­­­­­o­­­­­­­­­­­­­­­­­­r’­­­­­­­­­­­­­­­­­­s ­­­­­­­­­­­­­­­­­­s­­­­­­­­­­­­­­­­­­i­­­­­­­­­­­­­­­­­­s­­­­­­­­­­­­­­­­­­t­­­­­­­­­­­­­­­­­­e­­­­­­­­­­­­­­­­­­r­­­­­­­­­­­­­­­­­­-­­­­­­­­­­­­­­­­­­in-­­­­­­­­­­­­­­­­­­law ­­­­­­­­­­­­­­­­­­makes ­­­­­­­­­­­­­­­­­­$76 ­­­­­­­­­­­­­­­­­­hourly ­­­­­­­­­­­­­­­­­­on ­­­­­­­­­­­­­­­­­­the ­­­­­­­­­­­­­­­­­­computer. ­­­­­­­­­­­­­­­­­­She ­­­­­­­­­­­­­­­­­­has ­­­­­­­­­­­­­­­­­­been ­­­­­­­­­­­­­­­­­­without ­­­­­­­­­­­­­­­­­­work ­­­­­­­­­­­­­­­­­­for ­­­­­­­­­­­­­­­­­­six ­­­­­­­­­­­­­­­­­­months ­­­­­­­­­­­­­­­­­­but ­­­­­­­­­­­­­­­­­­last ­­­­­­­­­­­­­­­­­­month ­­­­­­­­­­­­­­­­­­her ­­­­­­­­­­­­­­­­­­income ­­­­­­­­­­­­­­­­­­was ­­­­­­­­­­­­­­­­­­$14418 ­­­­­­­­­­­­­­­­­­just ­­­­­­­­­­­­­­­­­­working ­­­­­­­­­­­­­­­­­­on ­­­­­­­­­­­­­­­­­­the ­­­­­­­­­­­­­­­­­­computer ­­­­­­­­­­­­­­­­­­for ­­­­­­­­­­­­­­­­­­a ­­­­­­­­­­­­­­­­­­fe­­­­­­­­­­­­­­­­­­w ­­­­­­­­­­­­­­­­­­ho­­­­­­­­­­­­­­­­­­urs. ­­­­­­­­­­­­­­­­­­vi­­­­­­­­­­­­­­­­­­sit ­­­­­­­­­­­­­­­­­­t­­­­­­­­­­­­­­­­­­h­­­­­­­­­­­­­­­­­­e ­­­­­­­­­­­­­­­­­­si­­­­­­­­­­­­­­­­­­te­­­­­­­­­­­­­­­­­­………….. ..www.J­­o­­­b­s­wo.C­­??
 

FormatC

Distinguished
Apr 4, 2011
981
1
18,990
All of the Radeon R9 290X cards are set to their Quiet Mode firmware setting
This is the jumper setting by factory and I've tested all cards out of the box in their default mode. I've measured the Tri-X dimensions and the results are listed in the table.
 
Status
Not open for further replies.