[Computerbase] R9 390 (X) Roundup

[RussianSensation]

http://www.computerbase.de/2015-10/neun-radeon-r9-390-x-partnerkarten-test/3/

Noise levels and temperatures

1. If you look at 5 benchmarks at 1440P, it's probably not worth the extra $ to get a 390X over the 390. Either both are > 60 fps or < 60 fps.

2. PowerColor PCS+ 390 and Sapphire Nitro are 2 stars of the 390 series, while Sapphire Tri-X 390X is probably the best 390X card, other than the PowerColor AIO CLC version.

Sapphire Nitro 390 = 38 dBA @ 70*C (Great result), total system power usage = 386W
PCS+ 390 = 36 dBA @ 76*C, total system power usage = 371W
Sapphire Tri-X 390X = 45.5 dBA @ 76*C = 430W

3. Asus continues on selling based on brand and not much engineering. It's the year 2015 and Asus is trying to put off-the-shelf GPU cooler on any random GPU with complete disregard for die sizes, die size orientation, heatpipe contact, etc. :whiste:

Asus Strix 390X/Fury

Asus Strix GTX980Ti - same garbage engineering

Asus R9 390X Strix = 51 dBA @ 89-91*C, total system power usage = 469-489W

Asus, congrats on making yet another

videocard heatsink design.

 

[tential]

There's a reason I am obssessed with Sapphire right now! Fury, R9 390, R9 290! All great coolers. Was surprised to see the r9 390x result though but it's still better than the competition. If I get an Arctic Islands card, I won't be surprised if it's a Sapphire Cooler that makes me do it.

 

[soccerballtux]

lol BUT WE HAVE 5 HEAT PIPES ignore that only 3 are making contact...

 

[Mondozei]

Yeah it's not just the normal fan noise either. On Prisjakt, basically a price aggregator in Sweden, a lot of people complained about coil whine on Asus GPUs for both GTX and Radeon cards in the GPU reviews.

Asus used to be awesome but they've since slipped pretty badly. The company has refocused on laptops, mobile phones and other such devices and it shows in their products.

People's perceptions have not yet caught up with that, but this is why this forum is so great.

 

[Gikaseixas]

The Strix cooler works better on NV high end I guess, I get around 73/74 degrees max on my card. The older ASUS DC2 design also had problems when bundled with 290/290x. I think these coolers are designed for Nvidia cards and adapted to AMD ones. That's only a theory

 

[3DVagabond]

lol BUT WE HAVE 5 HEAT PIPES ignore that only 3 are making contact...

In B4 the Asus fans start posting that it doesn't matter. Massive marketing FTW.

We've pointed this out before and some people went on and on that just having the pipes next to each other would transfer the heat just like they were in contact with the GPU.

 

[Seba]

We've pointed this out before and some people went on and on that just having the pipes next to each other would transfer the heat just like they were in contact with the GPU.

The heat pipes not in direct contact with the GPU will not transfer the same amount of heat, but they will transfer some heat (they are heated by the heat pipes next to them).

Did anyone proved otherwise? For example by removing those two extra heat pipes and comparing the GPU temperatures (before - with 5 heat pipes - and after - with the remaining 3 heat pipes)?

You should also note that the 3 heat pipes in direct contact with the GPU are dispersing the heat to the edges of the heatsink. Those not in direct contact have the other end closer to the GPU anyway. Just because some manufacturers are encasing all the heat pipes in a shiny coper block does not mean that other solutions do not work at all. Without the coper block encasing them, the heatpipes in direct contact with the GPU may even transfer the heat better. But it's not shiny, so it must be worse, right?

 

[3DVagabond]

They will hardly help at all because they remove heat from the pipe next to them only. They don't actually remove any heat from the GPU.

And nobody destroyed a heat sink just to prove something that is very obvious simply by looking at it.

Again, those pipes make no contact with the GPU so they won't remove heat from it. They will remove some heat from the heatpipes that are making contact, but the fins, which are actually designed to do that, will be far far more efficient. So the benefit will be minimal.

 

[LTC8K6]

It's probably cheaper to make one block that will fit a variety of die sizes and can handle future bigger die sizes?

 

[flopper]

asus needs to innovate as next year cards go Nano Nano small

 

[lehtv]

I was so hoping for Gigabyte R9 390 G1 Gaming to be in this test, but no. There's still not a single actual review of this card out there.

Also, PCS+ seems amazing. My next card might just be a PowerColor

 

[AtenRa]

They will hardly help at all because they remove heat from the pipe next to them only. They don't actually remove any heat from the GPU.

And nobody destroyed a heat sink just to prove something that is very obvious simply by looking at it.

Again, those pipes make no contact with the GPU so they won't remove heat from it. They will remove some heat from the heatpipes that are making contact, but the fins, which are actually designed to do that, will be far far more efficient. So the benefit will be minimal.

By conduction, heat is removed from the GPU by all (*) heat-pipes no matter if not all pipes touch the GPU directly.
We have heat transfer from the GPU to the first pipe (direct GPU contact) and then to the next pipe (pipe to pipe contact)
Obviously you transfer more heat at the same time from direct contact to the the GPU. The reason is the same as with CPU to heat-sink, losses from the contact between the two pipes.

* That is only if all pipes touch with each other in order to have heat flow from one to the other.

 

[Essence_of_War]

By conduction, heat is removed from the GPU by all (*) heat-pipes no matter if not all pipes touch the GPU directly.
We have heat transfer from the GPU to the first pipe (direct GPU contact) and then to the next pipe (pipe to pipe contact)
Obviously you transfer more heat at the same time from direct contact to the the GPU. The reason is the same as with CPU to heat-sink, losses from the contact between the two pipes.

* That is only if all pipes touch with each other in order to have heat flow from one to the other.

Notably absent from your assertion is any indication of the relative efficiency of conduction between heat pipes vs. conduction and phase transition within a single direct contact heat pipe.

You may be strictly correct that since there is SOME conduction between heat pipes that therefore all heat pipes cool the GPU, but my suspicion is that this claim is analogous to looking out the window of an airplane, seeing a jet engine and a balsa wood fan, and saying that the airplane's thrust is provided by both turbine engines and balsa wood fans.

 

[zentan]

Spoiler

http://www.computerbase.de/2015-10/neun-radeon-r9-390-x-partnerkarten-test/3/

Noise levels and temperatures

1. If you look at 5 benchmarks at 1440P, it's probably not worth the extra $ to get a 390X over the 390. Either both are > 60 fps or < 60 fps.

2. PowerColor PCS+ 390 and Sapphire Nitro are 2 stars of the 390 series, while Sapphire Tri-X 390X is probably the best 390X card, other than the PowerColor AIO CLC version.

Sapphire Nitro 390 = 38 dBA @ 70*C (Great result), total system power usage = 386W
PCS+ 390 = 36 dBA @ 76*C, total system power usage = 371W
Sapphire Tri-X 390X = 45.5 dBA @ 76*C = 430W

3. Asus continues on selling based on brand and not much engineering. It's the year 2015 and Asus is trying to put off-the-shelf GPU cooler on any random GPU with complete disregard for die sizes, die size orientation, heatpipe contact, etc. :whiste:

Asus Strix 390X/Fury

Asus Strix GTX980Ti - same garbage engineering

Asus R9 390X Strix = 51 dBA @ 89-91*C, total system power usage = 469-489W

Asus, congrats on making yet another

videocard heatsink design.

Nice round up.Sapphire has got it very very right for air starting from Tri-X 290/X.Both noise and thermals.

Asus didn't learn the lesson with their DC2 290/X.What is pathetic is that there are some salesmen on various forums promoting these cards as the very best of the 390 or 290 family with stupid brandworshipping comments just because they get some freebies or something.

 

[Seba]

So do you think that the heat pipe in direct contact with the GPU heats to 70C for instance, while the heat pipe next to it remains at 24C or something?

Your suspicion might be just that: a suspicion, not a proven fact.

 

[Head1985]

Asus just dont care about AMD.They make excelent coolers only for NV cards..and then they using same cooler from NV in AMD cards.

 

[Headfoot]

I dont know why we're acting like the ASUS approach could "theoretically" be ok. Look at the numbers. Its the loudest and hottest one of the bunch by a mile. Case closed.

 

[Seba]

I dont know why we're acting like the ASUS approach could "theoretically" be ok. Look at the numbers. Its the loudest and hottest one of the bunch by a mile. Case closed.

Case reopened.

Temperature differences could be caused by other factors (different radiant surface area, diferent spacing and position of the heatsink fins, different materials of the heatsink, different fans). Not necessarily because the heat pipes are not enclosed in a block. It could be this, but I did not see any demonstration of that (excluding the "it looks like it is worse" type of arguments).

 

[Stuka87]

The Strix cooler works better on NV high end I guess, I get around 73/74 degrees max on my card. The older ASUS DC2 design also had problems when bundled with 290/290x. I think these coolers are designed for Nvidia cards and adapted to AMD ones. That's only a theory

Has nothing to do with that, the cooler is a poor design. Testing on 980Ti's shows the same thing. There is huge gaps between the heat pipes, and half of them are nothing nothing at all. Its a poor design.

 

[Stuka87]

So do you think that the heat pipe in direct contact with the GPU heats to 70C for instance, while the heat pipe next to it remains at 24C or something?

Your suspicion might be just that: a suspicion, not a proven fact.

The ones next to the ones touching are sure to warm up some, but there is an air gap between them and the ones that are touching the GPU. So the ones not touching are not much to aid in the cooling of the GPU. Which means the heatsink side they are connected to is a waste.

Just look at temps from other GPUs that have better designs, their temps are all lower.

 

[LTC8K6]

Just point an IR camera at it and see if those pipes are doing anything.

I'll bet a donut that they are. :biggrin:

 

[Techhog]

The pipes are obviously doing something, but not as much as the ones making direct contact. On top of that, this is called a Direct CU cooler. It's borderline false advertising. It's also basically a quiet reference cooler, considering how hot it gets. It needs a revision, end of story.

 

[thesmokingman]

Better use of that Asus card... take a seat.

 

[Gikaseixas]

Has nothing to do with that, the cooler is a poor design. Testing on 980Ti's shows the same thing. There is huge gaps between the heat pipes, and half of them are nothing nothing at all. Its a poor design.

No, just no. I have the card and i tested it already, mine tops out at 74 degrees max and that's overclocked so i still believe that ASUS coolers work better on Nvidia cards.

https://www.techpowerup.com/reviews/ASUS/GTX_980_Ti_STRIX_Gaming/34.html

 

[crisium]

Guess the Tri-X 390X isn't as good as the Fury Tri-X. This thing is dead silent in an R4.