AMD R9 390X Coming With Cooler Master Liquid Cooler + Estimated Performance

[Opcode]

This thing better be fast with tdp like that.

You make 300w sound drastic... It's only 10w more than the R9 290X meanwhile offering 45% more performance.

 

Well seems like AMD is getting their crap together (somewhere). These new GPU's from AMD seem to need a lot power. Well that's the reason to buy bigger PS's. Is there a contract behind that? :huh:  I guess not^^

But I hope they do something(in the near future) about that big heat output. I hope AMD strikes back in every possible way. 

I want Nvidia to get their crap together and do something amazing. I hate them for the 3,5 Gigs. They made a big mistake with that.

I want a brutal GPU war.

If the R9 390X is rated at 300w then it's power consumption should be similar to the R9 290X. Basically a drop-in replacement for existing R9 290X owners.

[Fetzie]

This thing better be fast with tdp like that.

45% more stuff. Same TDP. Doesn't take a PhD in maths to figure out how much more efficient the 390X is compared to the 290X.

[patrickjp93]

Fins don't come in contact with the liquid. They connect to the main channel that runs through the radiator sort of like how heatpipes run through aluminum fins. The amount of copper used in an AIO is really not a whole lot (definitely less than $5 worth). The block core is thin with fins meanwhile the radiator probably uses copper tubing for it's channeling of water. You can buy copper tubing for like $1 a foot.

Any good radiator these days has the liquid flowing through the fins, but w/e. If cheap and ineffective is your thing...

 

One thing to grasp is GCN 1.3 may bring some major power improvements. How can the R9 390X be rated at 300w meanwhile having 20 more compute units than the R9 290X at 290w. That's an extra 1280 SP's crammed into an extra 10w. If that's the case then GCN 1.3 has to of seen some major power cuts to keep heat output down that much. Now the thing is AMD stacked on a bunch of extra compute units. So the card is probably going to draw around the same power as the R9 290X. Tho if they didn't and left the shader count at 2816 then the R9 290X equivalent would of been 206w. That's quite a difference at around a 30% improvement in TDP.

45.45% more cores 4096/2816 = 1.454545...

Sorry, I forgot compute units for AMD are like SMM/SMX for Nvidia.

[Opcode]

Any good radiator these days has the liquid flowing through the fins, but w/e. If cheap and ineffective is your thing...

All radiators pump the liquid through a channel connected to the fins hence why you can see directly through the fins and out the other side. The fins are only there to create surface area for heat dissipation. They work no different than the radiator found in your car. If you've ever tore one of them apart for scrap metal (I have) there is a maze of copper piping that runs through the aluminum fins. A step up from this would be using copper fins which are more conductive but come with added cost.

[patrickjp93]

All radiators pump the liquid through a channel connected to the fins hence why you can see directly through the fins and out the other side. The fins are only there to create surface area for heat dissipation. They work no different than the radiator found in your car. If you've ever tore one of them apart for scrap metal (I have) there is a maze of copper piping that runs through the aluminum fins. A step up from this would be using copper fins which are more conductive but come with added cost.

Older radiators have pipes which run through the array of fins, but the very best actually carry the water through the insides of the fins themselves. This is why sometimes if you break a fin you can cause a permanent leak.

[Khvarrioiren]

You can't have a copper block and aluminum fins. It would destroy the entire loop within a year.

 

Doesn't change the fact that every single Asetek radiator is 100% aluminium. Check their own specifications. For example their top-end cooler http://asetek.com/desktop/cpu-coolers/590lx.aspx

[NAP51DMustang]

You can't have a copper block and aluminum fins. It would destroy the entire loop within a year.

Nepton series are copper block and aluminum rad core. So, yes you can.

[patrickjp93]

Nepton series are copper block and aluminum rad core. So, yes you can.

Long term, no, you cannot. Get on my damn level. http://martinsliquidlab.org/2012/01/24/corrosion-explored/

[Swndlr]

i wanna sell my 770 and put money towards this

[Khvarrioiren]

Long term, no, you cannot. Get on my damn level. http://martinsliquidlab.org/2012/01/24/corrosion-explored/

 

Better contact Asetek and tell them to get on your level.

[NAP51DMustang]

Long term, no, you cannot. Get on my damn level. http://martinsliquidlab.org/2012/01/24/corrosion-explored/

Or I could just not as you aren't above me in any way? Especially seeing as how multiple AIO manufacturers use Aluminum rads and Copper blocks and those last years on end with no issues.

And for some variety in this discussion http://www.corsair.com/en-us/hydro-series-h110-280mm-extreme-performance-liquid-cpu-cooler:: Copper block, Aluminum rad, 5 year warranty.

[patrickjp93]

Or I could just not as you aren't above me in any way? Especially seeing as how multiple AIO manufacturers use Aluminum rads and Copper blocks and those last years on end with no issues.

And for some variety in this discussion http://www.corsair.com/en-us/hydro-series-h110-280mm-extreme-performance-liquid-cpu-cooler:: Copper block, Aluminum rad, 5 year warranty.

Cheap rad with fin-on-pipe design. The pipes could be PVC for all you know and it wouldn't cause a problem in reactions. In radiators where the fins are the pipes, such as in the GT/X black ice line, Alphacool Nexxos, or EK Coolstream, you'd never get away with it. Now, I am a liquid cooling enthusiast who doesn't settle for weak Asetek rebrand AIOs, but if you want to throw money away on a unit which is not only inferior in cooling but in noise and longevity while also not being versatile for future socket changes, be my guest. You mix aluminum and copper and it's a recipe for disaster, period, whether you lose the unit or not due to galvanic corrosion.

[patrickjp93]

Better contact Asetek and tell them to get on your level.

Asetek wouldn't exist if it weren't for the generation of enthusiasts I grew up with and competed against. Their copyrighted AIO designs were stolen from public competitions back in the early 2000s. Asetek is nowhere near my level. They just had enough money to buy government protection from competition and thusly the investors to get mass production up and running while patent trolling anyone who dares design competing products.

Their radiators are some of the worst in the business because of fin-on-pipe design and by using aluminum fins. If the internal pipe is copper, steel, or is at least nickel-plates, then your problems are probably averted as long as the only thing running through it is water and a little anti-freeze to keep away bacterial/algal growth. That said, you're guaranteed vastly inferior performance, and it shows.

[Opcode]

Long term, no, you cannot. Get on my damn level. http://martinsliquidlab.org/2012/01/24/corrosion-explored/

The source is kind of sketchy as he states that he filled them with water before he stored them away for a year. Who does that? With both of the radiators not being closed loops its pretty obvious a constant supply of both oxygen and water are readily available to both of the units (corrosion breeding ground). It was really his own stupid idea to fill a radiator with water for storage under them circumstances. If he drained them and stored them bone dry for a year they would still looked and functioned like new. AIO units don't face these same issues as they are sealed air tight one time during manufacturing. I can actually prove your theory wrong right now as I got an AIO closed loop in my closet that's been sitting there for two years (it has been cycled when I first got it). Still looks and functions like brand new.

 

Side of the radiator that has been sitting in this exact position for over a year (still like brand new).

Entire side of the rad (still like brand new).

Block with the factory compound gone (as I've installed it before) so it has been cycled (still like brand new w/o compound ).

 

Yeah... corrosion not happening on an AIO.

[handymanshandle]

So for more than a two way Crossfire set up how does that work and will that mean that there will be no different styles of coolers on it like the R9 295X2? 

I mean it sounds cool but I think I might stick to going with an R9 290(X) mostly because it's cheaper and there is more variety.

(That also means that the price might not reflect performance but rather the overall cost of the cooler... Though I suppose then I would have more motivation to make a custom loop) Maybe an option is that they have like a model that is the regular card but you have to get your own block and loop or they pack in an EK kit for you to throw on it. (Just some ideas.) 

[Winvsking]

GEEEZZZ.  That card may pull me to the red team

[patrickjp93]

The source is kind of sketchy as he states that he filled them with water before he stored them away for a year. Who does that? With both of the radiators not being closed loops its pretty obvious a constant supply of both oxygen and water are readily available to both of the units (corrosion breeding ground). It was really his own stupid idea to fill a radiator with water for storage under them circumstances. If he drained them and stored them bone dry for a year they would still looked and functioned like new. AIO units don't face these same issues as they are sealed air tight one time during manufacturing. I can actually prove your theory wrong right now as I got an AIO closed loop in my closet that's been sitting there for two years (it has been cycled when I first got it). Still looks and functions like brand new.

Side of the radiator that has been sitting in this exact position for over a year (still like brand new).

Entire side of the rad (still like brand new).

Block with the factory compound gone (as I've installed it before) so it has been cycled (still like brand new w/o compound ).

Yeah... corrosion not happening on an AIO.

BS. Give it a month to cycle and, if the internals contain both copper and aluminum, you'll find flakes circulating around, and then another month or two later the motor will be whirring, and then two or three later it'll be gone. No matter whether it was sealed or not, gallivanic corrosion does not need a constant supply of oxygen, and, like rusting, it only takes a single starting molecule for it to begin the chain reaction. Also, there's no need for him to store it bone dry and the short amount of time that the loop would be open is negligible to dissolve much air into the fluids. Your approach at debunking based on your very limited understanding of chemistry on multiple fronts is nothing short of fallacy.

[Opcode]

BS. Give it a month to cycle and, if the internals contain both copper and aluminum, you'll find flakes circulating around, and then another month or two later the motor will be whirring, and then two or three later it'll be gone. No matter whether it was sealed or not, gallivanic corrosion does not need a constant supply of oxygen, and, like rusting, it only takes a single starting molecule for it to begin the chain reaction. Also, there's no need for him to store it bone dry and the short amount of time that the loop would be open is negligible to dissolve much air into the fluids. Your approach at debunking based on your very limited understanding of chemistry on multiple fronts is nothing short of fallacy.

It's cycled plenty enough as it took nearly a month for them to send me more. Either way that's concrete evidence that corrosion in a closed loop is not likely to happen for an extended period of time. This unit has been sitting like it is for over two years now which is double the time period of your source (and has probably seen more abuse). Corrosion may be a threat to custom loops (which deal with oxidation) tho AIO units don't suffer from the same issues. I can probably bump this thread 5 years from now with new pictures of this same exact loop showing it looking and functioning like brand new. It'll break down in time but sure in hell not in just one year of sitting around.

[Khvarrioiren]

BS. Give it a month to cycle and, if the internals contain both copper and aluminum, you'll find flakes circulating around, and then another month or two later the motor will be whirring, and then two or three later it'll be gone. No matter whether it was sealed or not, gallivanic corrosion does not need a constant supply of oxygen, and, like rusting, it only takes a single starting molecule for it to begin the chain reaction. Also, there's no need for him to store it bone dry and the short amount of time that the loop would be open is negligible to dissolve much air into the fluids. Your approach at debunking based on your very limited understanding of chemistry on multiple fronts is nothing short of fallacy.

 

My friend has been running an Antec Kuhler H2O 620 for well over a year now, which has a copper heatplate and a full aluminium radiator, and the pump is as quiet as ever. Give your ego a rest for once in a while, and don't try to make your opinions into undisputable gospel of the internet.

[shawno]

Patrick...

Most radiators are copper with alloy fins because its cheap and works just as good. All copper radiators are really expensive and the difference is negligible

Also...im not sure what you mean by the expensive radiators have the coolant flowing through the fins... this would actually be worse for cooling and the fins act as a heat exchange for the veins.

This has been the same for many years and even car radiators use the same principal

[freemini]

The people that will be buying this card won't be doing it in a 30 dollar case with one fan in the front and one fan in the back.

What if those two fans are $30 fans? This is legitimately the first time I've seen someone argue that air cooling is better, usually it's all about the crazy water loops. I suppose this is why you can choose what design you want, this is PC after all.

[Guest Strangerbob]

you people are so used to marginal jumps between generations that you find this exiting?

 

i remember times when every generation was a jump like this or higher

 

 

like the first DX10 cards - 8800GTS, and 8800GTX....  at least TWICE the performance of their previous generation equivalents

[bogus]

Too bad you can't get it to the US or Europe and stay under $35 Of course you can get things cheaper in China. The question is if you can get it to you and stay under cost. You can't. Care to put in some actual effort?

 

Sure you can, you just have to know how to do it.

And like I said: this is a quick example, looking at the first page results and with no negotiation (or PO).

I'm sure that companys like CM, FD, etc can get this for less then 15$. Again, no where near 50$.

[Memphaestus]

Cheap rad with fin-on-pipe design. The pipes could be PVC for all you know and it wouldn't cause a problem in reactions. In radiators where the fins are the pipes, such as in the GT/X black ice line, Alphacool Nexxos, or EK Coolstream, you'd never get away with it. Now, I am a liquid cooling enthusiast who doesn't settle for weak Asetek rebrand AIOs, but if you want to throw money away on a unit which is not only inferior in cooling but in noise and longevity while also not being versatile for future socket changes, be my guest. You mix aluminum and copper and it's a recipe for disaster, period, whether you lose the unit or not due to galvanic corrosion.

Dude, the liquid does not flow through the .038- .2 mm thin fins in any of those designs.  Imagine how thin those channels would be, and how little pressure it would take to rupture those channels.  You clearly don't know what you are talking about. 

 

Galvanic Corrosion really isn't a problem with aluminum fin and copper pipe radiators because the water never touches the aluminum.  Household AC units have been mixing aluminum and copper even in the piping for forever.  Auto radiators have mixed aluminum and copper in their construction since their inception as well.  And every other radiator style heat exchanger has at one point or another mixed the two metals.  It really isn't as big of a deal as you are making it out to be.

[Gunjob]

Got just over £1000 saved so far readying myself for a Holiday and some 390x's in Q2. Honestly really excited to have AIO's the noise reduction will be welcome. I might even get a whole set of Noctua's to pair up with it. 

 

Edit:

 

@Strangerbob  Don't forget the jump from 4870 to 5870. I loveeeddd my 5870, hell of a card. http://www.hwcompare.com/1000/radeon-hd-4870-1gb-vs-radeon-hd-5870/