Radeon R9 290X Review: AMD's Back In Ultra-High-End Gaming

[smeezekitty]

Working link: http://www.legitreviews.com/nzxt-kraken-g10-gpu-water-cooler-review-on-an-amd-radeon-r9-290x_130344

I wonder why the clock speed did not go up.

 

[Gurg]

Working link: http://www.legitreviews.com/nzxt-kraken-g10-gpu-water-cooler-review-on-an-amd-radeon-r9-290x_130344

I wonder why the clock speed did not go up.

First they ran the test in a 70º F normal temp room rather than 78º like one tester, also they ran the reference at uber, so they found that it didn't throttle at 1080p. Evidently they are going to run another test batch and article with overclocks and hopefully at higher resolutions and more tests to push the card,

 

[Novuake]

Review of Kraken G10 bracket for closed loop coolers installed on 290x. 50ºC drop in load temps. 21% drop in noise levels.
http://www.legitreviews.com/nzxt-kraken-g10-gpu-water-c...

Page not found?
I doubt 50C drop is very accurate... Can not see that even being possible.

They reckon it is.

http://www.legitreviews.com/wp-content/uploads/2013/12/r9-290x-temperature-testing-645x531.png

And do you believe it?
Considering that 50C is like night and die on the surface of Mercury when it comes to CPU/GPU temps.
And that the R9 290X's target is best performance at 90+C.

 

[17seconds]

Review of Kraken G10 bracket for closed loop coolers installed on 290x. 50ºC drop in load temps. 21% drop in noise levels.
http://www.legitreviews.com/nzxt-kraken-g10-gpu-water-c...

Page not found?
I doubt 50C drop is very accurate... Can not see that even being possible.

They reckon it is.

http://www.legitreviews.com/wp-content/uploads/2013/12/r9-290x-temperature-testing-645x531.png

And do you believe it?
Considering that 50C is like night and die on the surface of Mercury when it comes to CPU/GPU temps.
And that the R9 290X's target is best performance at 90+C.

The most surprising thing about that review is they showed zero difference in performance. The GPU even continued to throttle below its minimum core clocks at that 50c temperature. That means all you're going to get with these highly anticipated custom coolers is less noise, not necessarily more performance. But we'll have to wait for the Part 2 Overclocking review to be posted.

 

[Novuake]

The most surprising thing about that review is they showed zero difference in performance. The GPU even continued to throttle below its minimum core clocks at that 50c temperature. That means all you're going to get with these highly anticipated custom coolers is less noise, not necessarily more performance. But we'll have to wait for the Part 2 Overclocking review to be posted.

Then i call BS on that review... Its not consistent with the Accelero review on Toms and that one makes perfect sense.

 

[Sakkura]

It's not BS. It's just that their reference cooled card doesn't throttle down, either because it's a golden sample or because of the benchmark sequence. An average of 991 or 992 MHz is extremely close to the 1000 MHz ceiling. Tom's easily pushed it down into the 800s.

 

[Mousemonkey]

Review of Kraken G10 bracket for closed loop coolers installed on 290x. 50ºC drop in load temps. 21% drop in noise levels.
http://www.legitreviews.com/nzxt-kraken-g10-gpu-water-c...

Page not found?
I doubt 50C drop is very accurate... Can not see that even being possible.

They reckon it is.

http://www.legitreviews.com/wp-content/uploads/2013/12/r9-290x-temperature-testing-645x531.png

And do you believe it?
Considering that 50C is like night and die on the surface of Mercury when it comes to CPU/GPU temps.
And that the R9 290X's target is best performance at 90+C.

I'm keeping an open mind at the moment, I use a CL cooler myself on my 560Ti so I am aware of their effectiveness and I'm sure other sites are going to pick this up and run their own tests so if this is BS it will soon be exposed as such won't it?

 

[Gurg]

It's not BS. It's just that their reference cooled card doesn't throttle down, either because it's a golden sample or because of the benchmark sequence. An average of 991 or 992 MHz is extremely close to the 1000 MHz ceiling. Tom's easily pushed it down into the 800s.

Tom's set their 290 to quiet slow fan mode not uber and was running the tests in a 78ºF hot room temps not a more realistic 70ºF as legit. A bunch of us mentioned the high room temps for the Toms tests. I can't imagine why Toms handicapped the card that way.

 

[Mousemonkey]

My room is currently 23c and can reach 25c easily so why is 21c "more" realistic?

 

[smeezekitty]

I don't think that a free degrees ambient will make a whole lot of difference anyway.
There are much more critical factors unless the room is very cold or very hot

 

[Someone Somewhere]

Also remember that the inside of your case is probably going to be warmer than your room is by a few degrees.

 

[ryanaza]

if the 260x also has XDMA, think i can crossfire it with my a10-6800k? The reason i ask is because this is going into my htpc and i have limited space. the 260x is the longest card i can physiaclly fit.

 

[Sakkura]

Doesn't matter, it wouldn't be worth it anyway. 384 shaders vs. 896 shaders is far too asymmetric.

 

[Gurg]

)

My room is currently 23c and can reach 25c easily so why is 21c "more" realistic?

My current temps in Denver
House thermostat 69ºF
CPU 19ºC H100 water cooled CPU high airflow case next to outside house wall
Motherboard 26ºC
GPU 28ºC

Guess I could put a space heater next to my computer to warm it to replicate the 78ºF test temps

Room temp in most business offices is probably 70-72ºF. At 78º worker dressed for business attire would be sweating and uncomfortable.

 

[logainofhades]

if the 260x also has XDMA, think i can crossfire it with my a10-6800k? The reason i ask is because this is going into my htpc and i have limited space. the 260x is the longest card i can physiaclly fit.

Crossfire would not be possible with that combo.

 

[xredex]

..so..where did they get a monitor that can handle 7680x1440? I want one =)

 

[Someone Somewhere]

3 2560x1440 screens in Eyefinity.

Also, very few office workers will have one of these, there's a couple of degrees between temps inside and outside the case, and once this thing had been going for a few minutes, your room does have a space heater.

 

[Antaredja Kartasasmita]

AMD just flipped over the market & Nvidia fanboys' head

 

[Wisecracker]

if the 260x also has XDMA, think i can crossfire it with my a10-6800k? The reason i ask is because this is going into my htpc and i have limited space. the 260x is the longest card i can physiaclly fit.

It might get interesting with a Kaveri APU if you have an FM2+ motherboard.

 

[jimmysmitty]

I Just wondering why does this card came with 32Gb gddr5 and see only 4Gb. The PCB show 16 Elpida EDW2032BBBG (2G each). This amount of memory will help a lot in large scenes wit Vray-RT.

Each of those packages has 2 Gigabit of memory. That adds up to 32 Gigabit. 8 bit equals 1 byte, so 32 Gigabit equals 4 Gigabyte.

Gigabit Sakkura? are you shure?
All the memory as i know is measure in Bytes not bits, bits is a measure of data transmission.
But maybe you right, after all it will make sense.
Thanks for replay.

This is absolutely correct. Take a 8GB DDR3 memory stick as an example. It has 16 memory packages normally. Each package is 4Gb (little b means bit, big B means byte) totaling 64Gb when divided by 8 means 8GB.

Everything in computers goes based on the bit system which is the base 8 system. It takes 8 bits to create 1 byte.

The GPU has 32Gigabit of RAM and 4GigaByte of RAM.

My room is currently 23c and can reach 25c easily so why is 21c "more" realistic?

Because people don't live in places like AZ where 21c costs a ton of money to keep a house at. I keep mine at 78F or 25C which in the room where the PCs are easily fluctuates between 25-30C thanks to our nice 100F summers.

 

[Mousemonkey]

I don't live in AZ.

 

[jimmysmitty]

I don't live in AZ.

I know that. My point is people don't realize that 21c is not realistic for everyone because not everyone lives in a place where it is easy to keep it that cool.

I don't for sure and most of the southern part of the US doesn't, especially during the summer.

 

[Mousemonkey]

Oh OK.

 

[Sakkura]

I don't live in AZ.

I know that. My point is people don't realize that 21c is not realistic for everyone because not everyone lives in a place where it is easy to keep it that cool.

I don't for sure and most of the southern part of the US doesn't, especially during the summer.

Whereas further north, and in much of Europe, 21C is more realistic than 25C.