Monday, January 7th 2019

ASUS to Introduce the ROG MATRIX GeForce RTX 2080 Ti Graphics Card With Integrated AIO "Infinity Loop"

ASUS is looking to one-up the competition in the flagship graphics card market with its ROG MATRIX rendition of the RTX 2080 Ti interpretatioin of NVIDIA's silicon. The ROG MATRIX GeForce RTX 2080 Ti will incorporate an AIO watercooling solution complete with waterblock, pump, radiator and fans in a triple-slot design. ASUS claims the Infinity Loop cooler will bring about performance equivalent of that of an external 240mm AIO - a bold claim, since physics and proximity of the radiator and fans themselves to the PCB and die should dictate a higher temperature than an otherwise completely external solution.

That said, the design of the Infinity Loop is likely one of the most interesting graphics card cooling designs in some time, with a completely self-contained solution. The shroud and side of the graphics card feature laser cutouts and LED-illuminated elements (because 2019 wouldn't be much different from 2018 in that regard). Out-of-the-box clocks are set at 1815 MHz before any additional overclocking, and the card takes power from 2x 8-pin power connectors. The cost? North of $1600. Now that's design, even in the price-tag.
Source: ASUS
Add your own comment

28 Comments on ASUS to Introduce the ROG MATRIX GeForce RTX 2080 Ti Graphics Card With Integrated AIO "Infinity Loop"

#2
Vayra86
Isn't every loop an infinity loop?

I know, I'm a simple guy, but...
Posted on Reply
#4
Vayra86
VulkanBros said:
....and infinity price....
888 dollars wouldn't be a bad thing :D
Posted on Reply
#5
E-curbi
The Terrible Puddle said:
This is stupid
It's only stupid if you're buying it.

If you're selling it - it's BRILLIANT! :roll:
Posted on Reply
#6
mouacyk
Integrated aio... VRMs still rely completely on passive cooling. This was a good opportunity to enhance the block to actively cool VRMs.
Posted on Reply
#7
dorsetknob
"YOUR RMA REQUEST IS CON-REFUSED"
wonder what the life of the pump is ?? MTF
Posted on Reply
#8
phanbuey
the VRM's aren't actually cooled, but rather 'baked' by this and most of these heatpipe coolers. The gpu pumps out a ton heat that then get's blasted into the VRMs; and also transfers to the vram chips. In my experience if you take the cooler off and just put a tiny fan there (a la the NZXT kracken) your cooling of those components actually improves considerably.

Also, an overloaded rad pumps out a TON of heat, esp if the pump is at a low speed - I've learned this by using the NZXT Kracken with a Corsair h55 on a 1080ti, and even with a 1600rpm fan the bottom of the rad was steaming hot, literally felt like the tubes were about to melt after an hour of gaming, and the air coming out of the rad could have slow roasted a ham.

I moved back to the heatpipe cooler since there was no way that h55 could take that kind of punishment regularly, so im curious to see how this does since it doesn't have a ton more surface area than a 120mm rad.
Posted on Reply
#9
Chloe Price
1600 usd card with some stupid AIO cooling? Custom loop is the only way for cards like that, if going for liquid.
Posted on Reply
#10
phanbuey
Chloe Price said:
1600 usd card with some stupid AIO cooling? Custom loop is the only way for cards like that, if going for liquid.
Don't these cards hit board power limit before temp anyways? Not sure a custom loop even worth it tbh.
Posted on Reply
#12
jabbadap
Afaik, last time I checked highest end Turing cards had very beefy and efficient VRMs, which should be fine even passive cooling up-to LN2/dice OC.

phanbuey said:
Don't these cards hit board power limit before temp anyways? Not sure a custom loop even worth it tbh.
Sure. Without shunt mods for by passing TDP limit, power is one limiting factor. But even without that limit, one would need subzero cooling to overcome that typical freq. wall.
Posted on Reply
#13
Jism
phanbuey said:
the VRM's aren't actually cooled, but rather 'baked' by this and most of these heatpipe coolers. The gpu pumps out a ton heat that then get's blasted into the VRMs; and also transfers to the vram chips. In my experience if you take the cooler off and just put a tiny fan there (a la the NZXT kracken) your cooling of those components actually improves considerably.

Also, an overloaded rad pumps out a TON of heat, esp if the pump is at a low speed - I've learned this by using the NZXT Kracken with a Corsair h55 on a 1080ti, and even with a 1600rpm fan the bottom of the rad was steaming hot, literally felt like the tubes were about to melt after an hour of gaming, and the air coming out of the rad could have slow roasted a ham.

I moved back to the heatpipe cooler since there was no way that h55 could take that kind of punishment regularly, so im curious to see how this does since it doesn't have a ton more surface area than a 120mm rad.
So explain to me if VRM's are designed to run up to 120 degrees, being monitored by sensors is such a bad thing. Maybe you should appy for another internet engineer and dictate on how VRM cooling should be done. Perhaps you can save the million of suffering VRM chips from being baked.
Posted on Reply
#14
gamerman
well, what is stupid and whats not.

i guarantie that that asus loop gpu offer much much more brain that amd vega 64,even vega is half cheaper.

amdvega still eat more juice but its also almost 50% slower.

buy that asus 2080 ti loop gpu and buy amdvega64,keep it 2 years, then try sell both. forget amdvega64. no1 buy it, but that asus goes at once and i promise, and good price.
Posted on Reply
#15
bogami
As I see it is the weight, because it fell and it got damaged. I do not know what ASUS wants to be something special, and in doing so, it's all bad.! Do they not get a subcontractor for making a block and maybe the price would be more justified. Profit is higher at a cheap $ 50 AIO of three slots fat brick! Well, you need a "weapon leaf" to buy this card! , not to mention , the sale of an internal body part !:laugh::kookoo::p.
Posted on Reply
#16
DR4G00N
phanbuey said:
Also, an overloaded rad pumps out a TON of heat, esp if the pump is at a low speed - I've learned this by using the NZXT Kracken with a Corsair h55 on a 1080ti, and even with a 1600rpm fan the bottom of the rad was steaming hot, literally felt like the tubes were about to melt after an hour of gaming, and the air coming out of the rad could have slow roasted a ham.
Well yeah of course it's hot, that waste heat from the GPU has to go somewhere. Your hand is not a good way to measure temperature btw, even 40-50C will feel too hot to touch for more than a few seconds. That AIO was doing the job that it was designed for.
Posted on Reply
#17
dorsetknob
"YOUR RMA REQUEST IS CON-REFUSED"
Such a fanism with no forgiving features

Oops nearly missed adding the quote tho i suspect most of us know i meant
gamerman said:
well, what is stupid and whats not.

i guarantie that that asus loop gpu offer much much more brain that amd vega 64,even vega is half cheaper.

amdvega still eat more juice but its also almost 50% slower.

buy that asus 2080 ti loop gpu and buy amdvega64,keep it 2 years, then try sell both. forget amdvega64. no1 buy it, but that asus goes at once and i promise, and good price.
Please forgive his incoherent command of the English/American language its probably not a native speaker
Posted on Reply
#18
Chloe Price
phanbuey said:
Don't these cards hit board power limit before temp anyways? Not sure a custom loop even worth it tbh.
My point is that having a card which alone costs more than a good gaming PC, shouldn't be ruined with something like a concept idea like this. Either good air cooling, or custom loop to add in the existing loop.
Posted on Reply
#19
phanbuey
Jism said:
So explain to me if VRM's are designed to run up to 120 degrees, being monitored by sensors is such a bad thing. Maybe you should appy for another internet engineer and dictate on how VRM cooling should be done. Perhaps you can save the million of suffering VRM chips from being baked.
Maybe you should learn how to read and spell. Maybe you would notice that I never said it was a bad thing.
Posted on Reply
#20
Robcostyle
What are they counting on? Completely ridiculous and useless product. And I’m not even talking about 1600 price tag - 1800-2500 us presidents, when product hit the shelves - like we don’t have enough of ngreedia beating a history record for single gpu price, these ones trying to charge even more - and by the way, it’s 1/3 of the price!!!

It’s about already mentioned before, inferior to fullcover cooler design, inherent to almost any AIO on the market. Cheap, FE compatible 2080Ti and watercool/aquacomp/EK fullcover will be always preffered by Me, myself and I (and have no doubt any custom loop user). Any other instance - some strix or !amp will do fine. And it will cost roughly 1100-1300$. Not even talking that FE+HQ fullcover is far more superior solution than these.......AIO’s.
VRM, PCB design won’t play better here - since pascal, You have gpu limit, voltage, current, temp, TDP...... every corner limited.

LN2, you might say? Well, one problem here aswell - it’s all tied up with sponsors and sponsorship, no sales for that part of market.

So, all in all - I have only one thing that comes to my mind - and it makes me curious “When PC market ended up being ‘savvy customers society , and became market of consuming society, with ridiculous overprice, lack of new ideas, and lazy ignorant developers, that want to eat much without working much?”

P.S. I am awared of ASUS policy, so I really,reaaally doubt they will unlock tdp limit like somekingpin/lightning does have.
Posted on Reply
#21
Blueberries
They should have just recycled the Poseidon with an AIO if they were going to go this route.
Posted on Reply
#22
Chaitanya
<div class="youtube-embed" data-id="9OzolCFqb4k"><img src="https://i.ytimg.com/vi/9OzolCFqb4k/hqdefault.jpg" /><div class="youtube-play"></div><a href="https://www.youtube.com/watch?v=9OzolCFqb4k" target="_blank" class="youtube-title"></a></div>
Posted on Reply
#23
Mysteoa
phanbuey said:
the VRM's aren't actually cooled, but rather 'baked' by this and most of these heatpipe coolers. The gpu pumps out a ton heat that then get's blasted into the VRMs; and also transfers to the vram chips. In my experience if you take the cooler off and just put a tiny fan there (a la the NZXT kracken) your cooling of those components actually improves considerably.

Also, an overloaded rad pumps out a TON of heat, esp if the pump is at a low speed - I've learned this by using the NZXT Kracken with a Corsair h55 on a 1080ti, and even with a 1600rpm fan the bottom of the rad was steaming hot, literally felt like the tubes were about to melt after an hour of gaming, and the air coming out of the rad could have slow roasted a ham.

I moved back to the heatpipe cooler since there was no way that h55 could take that kind of punishment regularly, so im curious to see how this does since it doesn't have a ton more surface area than a 120mm rad.
VRM have max safe temps up to around 125C, so as long a your gpu is not hotter then 100C and you are blasting it with air the VRM will be the fine.
Posted on Reply
#24
yakk
Interesting engineering, but for the probable price just a full block on an open loop sounds a whole lot more reasonable.
Posted on Reply
#25
cadaveca
My name is Dave
I see no one noticed the hardware support for volt mods built in the card; something the MATRIX series has had always. It's nice to see a true enthusiast-oriented PCB for GPUs here, but I wanna hear more about the volt mods!
Posted on Reply
Add your own comment