ASUS ROG Dominus Pictured, Core i9 XCC Confirmed to Feature 6-channel Memory

[silapakorn]

Finally a PC that can run PUBG without stuttering.

 

[Jism]

The 5Ghz model was demonstrated using a 1000W chiller. See https://www.tomshardware.com/news/intel-28-core-processor-5ghz-motherboard,37213.html

But they coud'nt actually do any demonstration since the power at that place was not sufficient to run both chiller & system, LOL.

 

[LiveOrDie]

This is what it looks like when Intel runs out of Ideas and can't be ass to design something new, a rebranded server chipset and CPU which needs a 3000000000w PSU to power it and a space vac to cool it. Look at all those fans, how retarded is that lol, let wait and see EK design a chipset water block for that those poor engineers lol.

 

[Valantar]

This is what it looks like when Intel runs out of Ideas and can't be ass to design something new, a rebranded server chipset and CPU which needs a 3000000000w PSU to power it and a space vac to cool it. Look at all those fans, how retarded is that lol, let wait and see EK design a chipset water block for that those poor engineers lol.

Well, I'd say managing a bench-stable 5GHz 28-core CPU is rather impressive in its own right, even if it's quite useless in practice. More of an engineering exercise than anything else. Still, commodities traders and futures traders will gobble these up at any cost, as more cores + higher frequencies = more transactions per seconds = more profits for them. And they can afford both $100'000++ PCs and replacing them if they fail regularly.

 

[LiveOrDie]

Well, I'd say managing a bench-stable 5GHz 28-core CPU is rather impressive in its own right, even if it's quite useless in practice. More of an engineering exercise than anything else. Still, commodities traders and futures traders will gobble these up at any cost, as more cores + higher frequencies = more transactions per seconds = more profits for them. And they can afford both $100'000++ PCs and replacing them if they fail regularly.

Yes but i think its just a tad un-developed it looks like Intel crapped there paint when AMD dropped there cracker size chip, so they have come out with this which is just retarted compared to AMDs, like so what its 28 cores AMDs is 32 cores and there boards are logical to have this thing is just a Pig.

 

[Valantar]

Yes but i think its just a tad un-developed it looks like Intel crapped there paint when AMD dropped there cracker size chip, so they have come out with this which is just retarted compared to AMDs, like so what its 28 cores AMDs is 32 cores and there boards are logical to have this thing is just a Pig.

Oh, no doubt, this is clearly a rush job. While Intels's demo motherboards were rather nicely made (if entirely oversized and very difficult to fit in a case), the ROG one looks like Asus had a couple of weeks to make the design work. At best.

Just like Z370 was rushed to get 6-cores into the mainstream to compete with AMD's 8 (leaving what is now called Z390 (which was originally supposed to be Z370 and launched half a year later) as some weird in-between unloved stepchild), this seems like Intel realized a couple of months ago that they were probably going to compete with 32-core Threadrippers later this year, so they took a server platform, made the bare minimum of changes to it required to call it a consumer platform, and pushed it out.

At least AMD was forward-thinking enough to use their server socket on X399, leaving them room to grow. They also have an "advantage" in not needing the gigantic VRM required to OC these chips, of course. OC'd TR might max out at 300-ish W (though likely somewhat higher for TR2) , while these likely run 700-1000W at max clocks depending on chip quality. As such, X399 can have more sensible motherboard designs. Not that the same isn't possible for this platform, but people will probably whine when they can't OC their $2-4000 CPU on their $1000 motherboards.

 

[XiGMAKiD]

What a great time for power user right now, from Intel you got 28C/56T monster with 6-channel memory and from AMD you got 32C/64T beast

The 5GHz Cinebench from Intel is kinda spoiling the fun though

 

[LiveOrDie]

Oh, no doubt, this is clearly a rush job. While Intels's demo motherboards were rather nicely made (if entirely oversized and very difficult to fit in a case), the ROG one looks like Asus had a couple of weeks to make the design work. At best.

Just like Z370 was rushed to get 6-cores into the mainstream to compete with AMD's 8 (leaving what is now called Z390 (which was originally supposed to be Z370 and launched half a year later) as some weird in-between unloved stepchild), this seems like Intel realized a couple of months ago that they were probably going to compete with 32-core Threadrippers later this year, so they took a server platform, made the bare minimum of changes to it required to call it a consumer platform, and pushed it out.

At least AMD was forward-thinking enough to use their server socket on X399, leaving them room to grow. They also have an "advantage" in not needing the gigantic VRM required to OC these chips, of course. OC'd TR might max out at 300-ish W (though likely somewhat higher for TR2) , while these likely run 700-1000W at max clocks depending on chip quality. As such, X399 can have more sensible motherboard designs. Not that the same isn't possible for this platform, but people will probably whine when they can't OC their $2-4000 CPU on their $1000 motherboards.

Indeed i just cant see this type of thing working not with that type of power needed, the power bill to play a game using that board would just be dumb lol.

 

[Valantar]

Indeed i just cant see this type of thing working not with that type of power needed, the power bill to play a game using that board would just be dumb lol.

These chips are not for gaming. Period. Even considering gaming as a purpose for these just shows how limited a scope most computer enthusiasts have for the usage of high-end PCs.

As for the viability, the power bill and power draw isn't an issue for anyone who can actually want/need the compute power (and can afford it). The issue is sub-ambient cooling, which isn't viable over any kind of time in pretty much any situation (unless you're running it in a hermetically sealed case stuffed with silica gel packets or some such zero-humidity environment).

 

[LiveOrDie]

These chips are not for gaming. Period. Even considering gaming as a purpose for these just shows how limited a scope most computer enthusiasts have for the usage of high-end PCs.

As for the viability, the power bill and power draw isn't an issue for anyone who can actually want/need the compute power (and can afford it). The issue is sub-ambient cooling, which isn't viable over any kind of time in pretty much any situation (unless you're running it in a hermetically sealed case stuffed with silica gel packets or some such zero-humidity environment).

I know but people see ROG and think gaming because it is a gaming line, also people do waste money on boards like this and only game on them, I brought a x7900 because i use the cores when rendering and programming, but even I wouldn't waste my cash on that things lol, to me i would feel like going backwards in motherboard engineering just because the thing is so hacked together.

 

[efikkan]

Maybe it is to you, because your budget could not afford to replace such a system should it die, but I know many who simply can't get enough, and a $50,000 machine is no big deal, and if it breaks, they'll just toss it and buy another. This market is actually quite large, but there aren't any products for this type of user as an enthusiast, so I am very happy to see these CPUs get a platform that has the clocking restrictions removed so I can sell such rigs and make 5k profit on a single PC.

Most (professional) users who basically want all the performance they can get do so because time is money to them. Even if they can afford replacing the CPU after a relatively short time, the time and work lost because of instability usually exceeds the minimal gains we get today from an overclock. You have to remember that a CPU will start creating corrupt data long before it causes many crashes, and since overclocking usually gradually degrades the stability of the system, it will be challenging to detect corrupted data before the system surpasses a certain threshold in instability, unless you keep running something like Prime95 to verify the system every week or so. I've seen first hand how much data loss and headache such instability can cause, so for me and many others overclocking a productive system is simply a no-go. Even using an overclocked system for just stability testing of software is challenging, it will result in a lot of random crashes which are not real bugs in the code, and spending resources on tracking down such potential problems can lead to a lot of wasted resources.