Sapphire's Vapor-X R9 290X 8GB: The More, The Merrier?

Status
Not open for further replies.

Janithdalw

Reputable
Nov 4, 2014
22
0
4,510
Very narrow difference in FPS between the R9 290X 8GB and R9 290X 4GB. The R9 290X 4GB is more than enough. More VRAM doesn't make much difference. I think Nvidia also released an 8GB version of the GTX 980. I would like to see how much FPS can the GTX 980 renders on 4K resolution. I guess even the GTX 980 8GB won't reach 60 FPS on 4k resolution.
 

bryanlarsen

Reputable
May 27, 2014
4
0
4,510
All the games you're testing are FPS games or have FPS style graphics. Why not also test other style of games, games that people are more likely to play at 4K? Civilization: Beyond Earth, for example.
 

JeanLuc

Distinguished
Oct 21, 2002
979
0
18,990
I'm not sure if I agree entirely with the conclusion. For me this is a win for the Saphire Vapour X 290x over a reference 290x (that's been overclocked to match the card it's going up against) rather then then a victory for 8gb over 4gb. The reason I say this is although the 290x is sound piece of tech reference coolers AMD choose to stick on the 290x are really poor and cause cards to throttle underload.

If 4Gb's wasn't enough then I would have expected to have seen the minimum fps nose dive, falling behind by 10% to 15% suggests the card wasn't boosting or was holding the clocks back to keep it from overheating. The only other explanation might be the choice of VRAM providers i.e some use Hynix others use Elpida and sometimes Samsung.
 

firefoxx04

Distinguished
Jan 23, 2009
1,371
1
19,660
stopped reading at "Sapphire is not considered a high end brand".

What are they supposed to do? They sell the best AMD has to offer with some of the best coolers on the market.
 
Not sure what difference it would ultimately make, but I suspect the sweet spot is 5760x1080 until 4K enters the mainstream (if ever). Good to know 4K will play, but Eyefinity is the logical path forward despite the push from 'early adopters' wanting 4k

Impressive Noise and Temps -- even if compared to the craptastic reference design. At first look though, does not look to be much of a change from OEM 'aftermarket' coolers on the 290X.

 

firefoxx04

Distinguished
Jan 23, 2009
1,371
1
19,660
No crossfire test? Umm.. the card by its self can barely manage 4k and thats when you beat the settings down far enough to where 8GB is no longer needed.

VRAM in crossfire is not 'doubled' when two cards are used so to compensate you need to have cards with more vram in the first place. Big oversight imo. Mine as well not even test the card by its self when we know 9 times out of 10 4GB is more than enough.
 

jdon

Honorable
Sep 30, 2013
34
0
10,540
Psst.... The paragraph describing the outputs says 280 instead of 290X...

[EDIT BY CLEVE] Thanks, fixed! [/EDIT]
 

airborn824

Honorable
Mar 3, 2013
226
0
10,690
Were is SoM with ultra textures in 1440p? that is the whole point of this card. compare 4gb models at 1080p the 8GB is for higher rez and eyefinity set ups. sometimes my bias is just is so obvious. I like both companies i just dont like how much digging i have to do to find truth.
 

mczak1

Honorable
Jul 23, 2012
11
0
10,510
Hmm I'd trust the numbers a lot more if the 4GB R290X wouldn't have been a reference card, which is well known to throttle left right and center due to the "not so good" cooling. Overclocking it isn't going to help matters there neither...
So I'd not be surprised if the alleged difference between the 4GB and 8GB would be due to cooling differences rather than actual memory amount difference...
 

Doug Lord

Honorable
Jan 8, 2014
73
0
10,630
This needs to be a Xfire contest in 4k @ MAX settings and custom texture packs. That's where 4GB MIGHT fall down. I know it's a very small segment of the population that runs this. Then I want to see the comparison to the Xone/PS4. I think people might find 90% is good enough when the option costs $5k.
 

Cryio

Distinguished
Oct 6, 2010
881
0
19,160
The whole point of 8 GBs and 4k is to make dual, tiple and quad setups.

290X in crossfire with 8 GBs would stomp these 970 in these games, if maxed.
290X in crossfire with 8 GBs would stomb even 980.

Tom's that an article we want to see.

290xs 8GBs vs 980s in 4k in games maxed out.
 
Interesting article. It still amazes me that now the r9 290X is $350 due to the 970 being $350 with r9 290X performance.

But back to topic, you guys never tested crossfire and SLI with those cards. I think there would be some great improvements to the r9 290X 8GB at 4k if it was CrossFired. Since 4k I would say still needs two high performance cards to be very enjoyable.

Even if I had the option for a radeon 8GB r9 290X, I think I would still go to the gtx 970 since that card really impressed me at what it can do at 4k, and mainly because both are pretty much equal even though the frame buffers are completely different sizes.
 

larkspur

Distinguished
"Sapphire isn't traditionally considered a premium brand..."

Don - You really can't make that statement without qualifying it. Gimmick LED lights and shiny magnesium alloy shrouds don't make a card "premium". Are you saying that Sapphire hasn't "traditionally" delivered high-performance and efficient cooling in their cards? Can you please explain your statement in more detail? Thanks!
 

anthony8989

Distinguished
%5 performance increase from %100 more VRAM vs the GTX 970 4GB even at 4k - nice. Put a GTX 980 4GB in the review and all of a sudden the 290x 8GB looks even more stupid. Tsk tsk AMD.
 
Error on the page 3 BF4 4K tests: it looks like you've swapped the labels for reference 290X and 970.

Speaking of the reference 290X, did you not have an aftermarket 4GB 290X available? If the idea is to isolate variables, why not get a 290X that wouldn't be hampered by poor cooling? Seems to me that would have been the better comparison.

And I too don't understand what you mean by Sapphire not being traditionally a premium brand. Tom's has given numerous awards to their cards in the past and the Vapor-X cooler has long been one of the best on the market.

And I have to agree with others here that 1440 and XFire tests would have been very appropriate for this card.

I'm not trying to tear you down, Don. I regularly enjoy your writing. I think you just missed the mark a little on this one.
 
Status
Not open for further replies.