Sapphire Radeon RX 5700 XT Pulse

[B-Real]

Especially when none of the cards result in better overclocks.
The Asus managed 30Mhz more clock speed over the worse card which is the Pulse and both of them are 15Mhz off from reference all overclocked.
Pretty much all within margin in clock speed. Silicon lottery is all there is.

In reality, Wizzard managed to pull nearly 4% OC with the Pulse and less than 1% with the Strix.

No not really if you look at other reviews of the Strix it's much cooler than the Pulse ! Not saying it's worth the money tho .

Of course there might be differences, but we don't know the setups: open bench, PC cases with side panel on, PC cases with side panel off. But at TPU, with the same setup, there was 2C difference in favor of Sapphire. If there will be a test site reviewing both cards, we will see. BTW, check Gamers Nexus Steve's review: with a 40 dB setting, it cooled the Pulse to 67C. And that 40 dB is a reference 1080Ti sound. Anyway, as you said, even if the Strix was cooler by 4-5C, it isn't worth the extra $90.

 

[Zubasa]

In reality, Wizzard managed to pull nearly 4% OC with the Pulse and less than 1% with the Strix.

Well the factory OC on the pulse is lower than that of the Strix.
In the end they all ended up practically the same.
In fact the Reference design used Samsung GDDR6 memory which in general OC better than Micron use in these AIB cards if not for the driver limit / bug.

 

[HD64G]

So, Sapphire delivered once again by providing a solid GPU, much better in thermals and acoustics than the ref one, while practically not being more expensive. Thanks for the review @W1zzard !

 

[Deleted member 158293]

Probably one of the better 5700XT models to be available. Looks like a great efficient cooler. Negatives I can see is memory lack of overclock, and the biggest negative being a 2.5 slot card instead of 2 slot.

On my shortlist.

 

[RH92]

Of course there might be differences, but we don't know the setups: open bench, PC cases with side panel on, PC cases with side panel off.

I was talking about Guru3D review so both used open bench . Strix hits 68°C in their review and 76°C in quiet mod that's a massive difference compared to TPU review .

But at TPU, with the same setup, there was 2C difference in favor of Sapphire. If there will be a test site reviewing both cards, we will see.

Wich makes litle sense considering the big difference in heatsink size and 1 more fan in favor of the Strix , unless there is something wrong with TPU Strix model ( or the Strix model on Navi in general ) that card should perform much better than the Pulse ! For sure im waiting to see if Guru3D has a Pulse review but yeah TPU results seem a bit odd .

 

[sutyi]

No not really if you look at other reviews of the Strix it's much cooler than the Pulse ! Not saying it's worth the money tho .

Do other sites have have both cards as well? Unless the same test bench and roughly the same ambient temperature is used it will skew results.

 

[RH92]

Do other sites have have both cards as well? Unless the same test bench and roughly the same ambient temperature is used it will skew results.

Guru3D has not a Pulse review yet i hope they publish one soon . I know ambient temps can skew results but both sites use open test bench . Guru3D is testing with 21°C ambient but i can't find TPU ambient .

 

[raptori]

Well done Sapphire , such card can steer the direction towards AMD and I'll definitely consider it as my next upgrade if I can get it near $410.

 

[Xuper]

You didn't use SoftPowerPlay mod for OC , Why ?

 

[jabbadap]

Well the factory OC on the pulse is lower than that of the Strix.
In the end they all ended up practically the same.
In fact the Reference design used Samsung GDDR6 memory which in general OC better than Micron use in these AIB cards if not for the driver limit / bug.

Speaking of it: Is there any third party OC software out yet to support overclocking navi cards? Wattman's 950MHz limit is low for gddr6, i.e. every nvidia card with 14Gbps micron tested here on tpu can do higher than that.

 

[ExplodingCaps]

Woah, it's very unlikely that asus strix can be beaten even with other 3 fans cooler let alone 2 fans card in term of thermal performance in rtx series. Pulse is really nailed it.

 

[cucker tarlson]

all in all strix oc vs pulse oc,strix wins.less oc heardroom but more out of the box performance for asus card.

compared to 2070 Super trio, 2070s is 16,4% faster than pulse and 15% faster than strix when comparing OC vs OC.

 

[Lindatje]

all in all strix oc vs pulse oc,strix wins.less oc heardroom but more out of the box performance for asus card.

compared to 2070 Super trio, 2070s is 16,4% faster than pulse and 15% faster than strix when comparing OC vs OC.

Getting on with OC and then asking for problems. Just insert the card, install the driver and start playing. You don't have to do it for 2 or 3 FPS more, and more problems.
I don't understand why people still want to overclock for that little bit more that you don't notice.

 

[cucker tarlson]

Getting on with OC and then asking for problems. Just insert the card, install the driver and start playing. You don't have to do it for 2 or 3 FPS more, and more problems.

oc is pretty effortless though.

 

[W1zzard]

Not really convinced by temperature charts . Normal BIOS mod 75°C / 35dBA , Quiet BIOS mod 74°C / 32dBA ? Unless there is massive variance between runs there is no way a card is cooler when fans spin slower !

it's explained in the text, two times

 

[Joss]

I like one of the cooling options on this card: the aluminium fins are parallel to the length of the card and both the rear and the front are open enough for air to flow out.
@W1zzard I think you should add a pic of the front to emphasise this. I had to confirm it with this video (@ 3:26):

 

[Turmania]

If Pulse series nailed it than watch out for the Nitro+ series. I'm waiting for the MSI Evoke review which is suppose to be reviewed by TPU soon I believe.

 

[RH92]

it's explained in the text, two times

Roger although as far as im aware 30mV difference is not enough to explain better temps with slower fans , let alone the marginal difference in frequency wich should have even less impact than the voltage . Don't get me wrong im not judging your work here it's just that those results seem a bit odd especially when compared to the Strix !

It would be nice to know what the ambient temps where during testing ( unless i missed this aswell ) .

 

[medi01]

Hm, Strix OCed higher, but consumes 20w less than Pulse, am I reading it right?

 

[jinxjx]

Once again why this wasnt tested on Ryzen cpus too ?

 

[SIGSEGV]

unigine haven? really?

/sad

 

[Zubasa]

Hm, Strix OCed higher, but consumes 20w less than Pulse, am I reading it right?

The Strix has much more efficient and overkill VRMs on the PCB.

 

[cucker tarlson]

Hm, Strix OCed higher, but consumes 20w less than Pulse, am I reading it right?

first time in a long time I'm not impressed with a sapphire product.Hope they have nitro coming,although that'd probably touch on the $500 price mark if it comes.

 

[RH92]

The Strix has much more efficient and overkill VRMs on the PCB.

That could also be explained by better binning on the Strix but indeed Strix has much more efficient VRM and runs cooler wich can easely explain the difference in both max clock and lower consumption ( despite using more fans ) .

 

[medi01]

That could also be explained by better binning on the Strix but indeed Strix has much more efficient VRM and runs cooler wich can easely explain the difference in both max clock and lower consumption ( despite using more fans ) .

That being said, in computerbase review, Strix consumes 30w more than pulse (and Pulse beats 2070S with A-bios, only +11w with S-Bios)