MSI Radeon R9 290X GAMING 4G Video Card Review @ [H]

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,601
MSI Radeon R9 290X GAMING 4G Video Card Review - On our test bench today is MSI's newest high-end GAMING series graphics cards in the form of the MSI Radeon R9 290X GAMING 4G video card. We will strap it to our test bench and compare it to the MSI GeForce GTX 780 Ti GAMING 3G card out-of-box and overclocked to determine which card provides the best gameplay experience.
 
I'm sorry if I missed it, but I keep looking and asking on all these cards for the R9 series.

Is this a reference layout? I know most custom cooling and factory OC's come with some sort of revision, but it would be nice to know for sure, since I would be putting on a waterblock anyway. All the reference cards prices' are still sky high but the 'custom' cards seem to have the lower pricing; on newegg at least. (which I find strange)

I just need another 290 so I can put a WB on it and crossfire, but 1: I don't want to pay the price for the reference cards and 2: I'm afraid of the 'custom' cards not being reference and the WB not fitting.

Any insight?
 
The 780Ti card has almost 25% more performance that the 290x in this review, if I remember correctly, these cards used to be evenly matched in most games, is this down to better Nvidia drivers ?
 
The 780Ti card has almost 25% more performance that the 290x in this review, if I remember correctly, these cards used to be evenly matched in most games, is this down to better Nvidia drivers ?
At reference speeds this is true, but I think the 780 Ti (or indeed Kepler in general) has always responded to overclocking better than the offerings from AMD. It is true that there have been some gains in NVIDIA's drivers, but I don't think they tell the whole story here. I hope NVIDIA isn't pulling some shenanigans in the image quality like they have in the past...
 
I hope NVIDIA isn't pulling some shenanigans in the image quality like they have in the past...
You say that as if AMD is spotless in that regard. :D

LancerVI said:
Is this a reference layout?
EK's configurator says it's not, larger capacitors.


Thanks for the review great as ever, shed some light on what I should look for.
 
Soooooo per this review the 780ti is a comparative bargain, interesting.
 
I'm sorry if I missed it, but I keep looking and asking on all these cards for the R9 series.

Is this a reference layout? I know most custom cooling and factory OC's come with some sort of revision, but it would be nice to know for sure, since I would be putting on a waterblock anyway. All the reference cards prices' are still sky high but the 'custom' cards seem to have the lower pricing; on newegg at least. (which I find strange)

I just need another 290 so I can put a WB on it and crossfire, but 1: I don't want to pay the price for the reference cards and 2: I'm afraid of the 'custom' cards not being reference and the WB not fitting.

Any insight?

No personal insight - I don't have a reference one handy to compare it to, though AlexisRO says no...

The 780Ti card has almost 25% more performance that the 290x in this review, if I remember correctly, these cards used to be evenly matched in most games, is this down to better Nvidia drivers ?

The card we compared it to has a ~20% factory overclock which accounts for the performance difference. If the 780 Ti was at reference clocks then you'd see a performance parity. There's really no reason to compare to a reference one when a 20% OC flavor available for about the same price :).
 
You say that as if AMD is spotless in that regard. :D


EK's configurator says it's not, larger capacitors.


Thanks for the review great as ever, shed some light on what I should look for.

No personal insight - I don't have a reference one handy to compare it to, though AlexisRO says no...


Thanks fellas. I appreciate it. Didn't realize EK had a configurator. Pretty much a Koolance guy here across the board, since they're down the street from me and I can pick up parts.

I'm still kicking myself for not just picking up two cards at launch. Now with mining, crazy prices. Still mad at myself! :mad: Starting to come back to earth a bit, but still!

Anyway thanks again.
 
Cards at launch were running at 90C and throttling speeds badly. I think everyone was in the same boat of wanting to wait for non-reference cards, and got kicked in the balls by the mining craze.
 
I hope the price comes down on this at 649 it's just not worth it. I'm in the market for a new card and I was hoping to get one of these MSI cards but not that this price point. The MSI 290 Gaming model is looking a lot better at $469. I think this card at 569 would be a better price point given the position of the 780 Ti.
 
Professional review as always. Much thanks.

On page one: "It is sporting one DVI-I, one DVI-D, one HDMI v1.4a and one DisplayPort connector." But the pics in article and MSI's site say two DVI-D connectors so no analog out.

Not a huge deal but there might be 1 in 100 people that plan to hook up an analog monitor.
 
The R9 290X perf numbers have regressed significantly. this review is riddled with inconsistency in scores when compared to the MSI GTX 780 Ti Gaming review. Either the MSI R9 290X sample is faulty or some other error. MSI GTX 780 Ti Gaming OC numbers remain the same or slightly higher as in the MSI GTX 780 Gaming review while the MSI R9 290X scores have all decreased when compared to the ASUS R9 290X DC2 which was used in the same MSI GTX 780 Ti Gaming review. The MSI R9 290X overclock was 1130 mhz while the ASUS R9 290X was 1115 mhz. Still the MSI R9 290X OC (1130 mhz) ends up significantly slower than the ASUS R9 290X(1115 Mhz).

http://hardocp.com/article/2014/02/10/msi_geforce_gtx_780_ti_gaming_3g_video_card_review/7

From the MSI GTX 780 Ti Gaming review

Crysis 3 High settings SMAA 4x
MSI GTX 780 Ti Gaming OC -55.7 fps
ASUS R9 290X (1115 mhz / 56700 Mhz) - 49.6 fps

From this review
MSI R9 290X(1130 Mhz) - 42.2 fps

http://www.hardocp.com/article/2014/02/24/xfx_r9_290x_double_dissipation_overclocking_review/5

From the XFX R9 290X OC review
XFX R9 290X (1130 mhz / 5800 Mhz) - 52.0 fps

the XFX R9 290X and ASUS R9 290X are consistent and the MSI R9 290X is not. BF4 lost 15% at Ultra MSAA 2x and roughly 6% at Ultra MSAA . This is in spite of ASUS R9 290X at 1115 Mhz.

BF4 Ultra MSAA 2x
ASUS R9 290X (1115 Mhz) - 72.3 fps
MSI R9 290X(1130 Mhz) - 63 fps

BF4 Ultra MSAA 2x
ASUS R9 290X (1115 Mhz) - 59.4 fps
MSI R9 290X(1130 Mhz) - 56 fps
 
Last edited:
Cards at launch were running at 90C and throttling speeds badly. I think everyone was in the same boat of wanting to wait for non-reference cards, and got kicked in the balls by the mining craze.

Yeah, but as I was going to put a waterblock on anyway, those problems were never an issue for me. I always want reference and will usually never buy a non-reference layout for that reason. Then I can do the OC's / Custom BIOS myself. Heat and throttling are never an issue.
 
$649 is too much--and $700 for the 780 is too much, too. Way too much. It's getting ridiculous.

At the original AMD suggested retail of ~$550 the design was at least a bargain compared to the $699 Ti. Now, like the 780, it's just too much. Much, too much...;)

I'll pass.
 
Good review thanks! notbad for a $100 premium, though I'm not a fan of MSI after my 280 Gaming fans died less than 2 months after purchase( running 60% too)


Found this typo:

The MSI Radeon R9 290X GAMING 4G comes equipped with a factory GPU clock of up to 1030MHz, a 30MHz increase over stock Radeon R9 290X GPU and features MSI’s Twin Frozr IV cooling system. It is configured with 4GB GDR5 memory running at a 6.0GHz effective rate.
 
The R9 290X perf numbers have regressed significantly. this review is riddled with inconsistency in scores when compared to the MSI GTX 780 Ti Gaming review. Either the MSI R9 290X sample is faulty or some other error. MSI GTX 780 Ti Gaming OC numbers remain the same or slightly higher as in the MSI GTX 780 Gaming review while the MSI R9 290X scores have all decreased when compared to the ASUS R9 290X DC2 which was used in the same MSI GTX 780 Ti Gaming review. The MSI R9 290X overclock was 1130 mhz while the ASUS R9 290X was 1115 mhz. Still the MSI R9 290X OC (1130 mhz) ends up significantly slower than the ASUS R9 290X(1115 Mhz).

http://hardocp.com/article/2014/02/10/msi_geforce_gtx_780_ti_gaming_3g_video_card_review/7

From the MSI GTX 780 Ti Gaming review

Crysis 3 High settings SMAA 4x
MSI GTX 780 Ti Gaming OC -55.7 fps
ASUS R9 290X (1115 mhz / 56700 Mhz) - 49.6 fps

From this review
MSI R9 290X(1130 Mhz) - 42.2 fps

http://www.hardocp.com/article/2014/02/24/xfx_r9_290x_double_dissipation_overclocking_review/5

From the XFX R9 290X OC review
XFX R9 290X (1130 mhz / 5800 Mhz) - 52.0 fps

the XFX R9 290X and ASUS R9 290X are consistent and the MSI R9 290X is not. BF4 lost 15% at Ultra MSAA 2x and roughly 6% at Ultra MSAA . This is in spite of ASUS R9 290X at 1115 Mhz.

BF4 Ultra MSAA 2x
ASUS R9 290X (1115 Mhz) - 72.3 fps
MSI R9 290X(1130 Mhz) - 63 fps

BF4 Ultra MSAA 2x
ASUS R9 290X (1115 Mhz) - 59.4 fps
MSI R9 290X(1130 Mhz) - 56 fps
If you look at the apples-to-apples graphs running at Very High system spec the numbers are about the same.

But my guess would be that DirectX performance has taken a hit in AMD's latest drivers. There have also been a few patches for BF4 that have been released since the time period of the other reviews.
  • XFX 290X DD Review: 14.1 Beta 1.6
  • MSI GTX 780 Ti GAMING 3G Review: 13.12 WHQL
  • MSI 290X GAMING 4G Review: 14.2 Beta
 
Edited

Even though your temps are well within normal operating limits, is there a chance your 290x could be throttling? Users elsewhere have reported throttling issues with the drivers you used, regardless of operating temperatures.

Especially given this result (OC is slower than non-overclock):
1396151094av674gYKyI_4_3.png
 
Last edited:
Edited

Even though your temps are well within normal operating limits, is there a chance your 290x could be throttling? Users elsewhere have reported throttling issues with the drivers you used, regardless of operating temperatures.

Especially given this result (OC is slower than non-overclock):
1396151094av674gYKyI_4_3.png

The throttling issue is a bug verified by AMD. ALL R9 290X cards throttle considerably when voltage and powertune (power limit) is pushed. On my R9 290X Tr-X even using +25mV causes my OC to fail.

13.12 driver = 1220MHz Core with +156mV and +50% power
14.x Beta = ~900Mhz the same voltage and power limit

This may explain why the overclock is so poor in this review and why the card is actually slower than stock in some cases. This does not negate this review in any way, it is after all AMDs fault for having buggy power tune/power limit in these beta drivers. Reviewers and users can only use the drivers available. If you want Mantle you must use 14.3 Beta, you just have to accept your OC is going to be severely limited. If you aren't using Mantle games then stick with 13.12 as the voltage and power limits work flawlessly.

Proof that AMD are aware of this problem.
http://forums.overclockers.co.uk/showpost.php?p=26072736&postcount=16

Watch the core clocks drop during the Tomb Raider bench. This is with +100mV and 50% Power Limit set. The more voltage the bigger the throttling.
http://www.youtube.com/watch?v=tTbxwKFI6No
 
It doesn't mean it's happening here though. OC results are higher in every other benchmark (indicating there is no throttling going on), and it could be that he just got the numbers swapped in this particular graph.
 
Currently in the process of moving a couple of states away, so I don't have access to my notes/data at the moment.

Edited

Even though your temps are well within normal operating limits, is there a chance your 290x could be throttling? Users elsewhere have reported throttling issues with the drivers you used, regardless of operating temperatures.

Especially given this result (OC is slower than non-overclock):
[snip]http://www.hardocp.com/images/articles/1396151094av674gYKyI_4_3.png[/img]

The card did NOT throttle and OC performance at +25mV as well as +40mV was much lousier than it was at +50mV. With regards to the BF4 numbers, I suspect I put a couple numbers in the wrong cell. I was wrestling with that chart for about an hour before I got it right (adding a 4th card to that was NOT fun), so I suspect I put the wrong number in. The OC'ed performance did give a better average FPS and I'll go back through my data and notes once the movers deliver it to my new house (read: might be a bit).

It doesn't mean it's happening here though. OC results are higher in every other benchmark (indicating there is no throttling going on), and it could be that he just got the numbers swapped in this particular graph.

I suspect a number swap on this one - everything is in boxes right now, so I'll double check it when I get my rigs set back up.
 
Nvidia does seem to hold back a bit on their clocks when it comes to reference hence why they overclock better on avg, Where as AMD seems to push all their chips pretty close to most they can do.

You say that as if AMD is spotless in that regard. :D

On the image trickery could ask that right now about AMD on BF4, some screen shots on mantle vs DX11, seems to be less draw distrance on mantle side. if that was fixed or not i don't know haven't heard anything on it.
 
Back
Top