This is a split board - You can return to the Split List for other boards.

290x: first single-gpu setup that should be recommended to have 650+ watts?

#21ShebeskiiPosted 10/26/2013 7:03:57 PM(edited)
DV8ingSources posted...
Xeeh is a terrible fanboy though. At least snuckie tends to back up his claims even if the proof sometimes appears to be based on outliers. When Xeeh argues, there is often misreading or misinformation involved.

Shebeski is a better green knight but he is more reasonable.

I'm team nVidia myself but I respect what amd has on offer and its the better objective choice in many cases.


I wouldn't use the word objective. AMD can be a better choice based on what you value, there's no arguing that. Some people like to make out Nvidia buyers as fanboys because they have different values. The irony is, that's fanboyism to judge people like that. So what if the price to performance is better? Does everyone buy a product based on one performance metric? No.

The 290X seems to be a tremendous value, especially for high resolutions, and will drop the price of Nvidia cards. Awesome. But the choice is usually much more ambiguous. In this case I feel it's an easy answer, but given the past (frame latencies, CF, CF profile support, general support and functionality of features such as V sync and multimonitor) it always hasn't.

Kudos to AMD for really bringing some serious competition. I may jump fence and come home. I am, after all, a previous AMD user (9600XT, X1900GT and 4890 owner here), and were my first serious introduction to PC gaming. I'm excited to see an AMD that's kicking ass without any BS to hold them back. Let's hope 2014 to be an awesome year to restore faith to the brand in the hive mind of PC gaming.
---
That which can be asserted without evidence, can be dismissed without evidence. - Christopher Hitchens
#22DarkZV2Beta(Topic Creator)Posted 10/26/2013 7:03:58 PM
Thescyy posted...
DarkZV2Beta posted...
Snuckie7 posted...
1300MHz runs at 1.3V? Yeah poke around on OCN a bit and you'll see. GK110 is a huge, power hungry chip. The only reason it doesn't consume much at stock is because it's clocked conservatively.


Got any links?


My computer would power off on a seasonic 850w bronze psu when I would bench with both my gpu's at 1255mhz+ along with my cpu at 4.8Ghz+. It would pull 1000w from the wall and power down about 5 seconds into the combined test every time.


That's not exactly 500 per card though, as was claimed of Titans. It is pretty surprising, though.
Did TFIII have a nonreference design?
---
Want that Shield!
Ball and Cup on ps mobile has framerate issues. -stargazer64
#23ThescyyPosted 10/26/2013 7:32:51 PM(edited)
DarkZV2Beta posted...
Thescyy posted...
DarkZV2Beta posted...
Snuckie7 posted...
1300MHz runs at 1.3V? Yeah poke around on OCN a bit and you'll see. GK110 is a huge, power hungry chip. The only reason it doesn't consume much at stock is because it's clocked conservatively.


Got any links?


My computer would power off on a seasonic 850w bronze psu when I would bench with both my gpu's at 1255mhz+ along with my cpu at 4.8Ghz+. It would pull 1000w from the wall and power down about 5 seconds into the combined test every time.


That's not exactly 500 per card though, as was claimed of Titans. It is pretty surprising, though.
Did TFIII have a nonreference design?


Well my killawatt was only up to 1kw measurement and I could see it having been more. That's also not watercooled, If I watercooled and push my volts passed 1.25v I'm sure it would get far worse because I hit a voltage wall around 1320mhz. I would personally say Snuckies arguement of 500w is a bit of an exaggeration though as it takes the furthest of overclocks to reach that, the average overclock is likely around ~400-450w.
---
i7 3930k @ 4.5Ghz | H220 | asrock x79 extreme 9 | 16GB 1600mhz ram | SLI GTX 780 | Samsung 830 256GB SSD | 8TB HDDs | EVGA Supernova 1300 G2 | Fractal design R4
#24Snuckie7Posted 10/26/2013 7:17:46 PM
DarkZV2Beta posted...
Thescyy posted...
DarkZV2Beta posted...
Snuckie7 posted...
1300MHz runs at 1.3V? Yeah poke around on OCN a bit and you'll see. GK110 is a huge, power hungry chip. The only reason it doesn't consume much at stock is because it's clocked conservatively.


Got any links?


My computer would power off on a seasonic 850w bronze psu when I would bench with both my gpu's at 1255mhz+ along with my cpu at 4.8Ghz+. It would pull 1000w from the wall and power down about 5 seconds into the combined test every time.


That's not exactly 500 per card though, as was claimed of Titans. It is pretty surprising, though.
Did TFIII have a nonreference design?

A GTX 780 at 1255MHz and 1.24V isn't a 1300MHz Titan at 1.3V either.

I forget where I saw the tests, but you'll see plenty of Titan owners claiming that their cards can pull 500W each, which isn't all that unreasonable if you extrapolate the power consumption from Thescyy's rig.
---
Intel Core i7 3820 | EVGA X79 SLI K2 | MSI 7950 Twin Frozr III | Samsung / 840 120GB / 8GB RAM | 1TB WD Caviar Blue | Corsair / 550D / H70 | Silencer MKIII 600W
#25DarkZV2Beta(Topic Creator)Posted 10/26/2013 7:28:12 PM(edited)
Snuckie7 posted...
DarkZV2Beta posted...
Thescyy posted...
DarkZV2Beta posted...
Snuckie7 posted...
1300MHz runs at 1.3V? Yeah poke around on OCN a bit and you'll see. GK110 is a huge, power hungry chip. The only reason it doesn't consume much at stock is because it's clocked conservatively.


Got any links?


My computer would power off on a seasonic 850w bronze psu when I would bench with both my gpu's at 1255mhz+ along with my cpu at 4.8Ghz+. It would pull 1000w from the wall and power down about 5 seconds into the combined test every time.


That's not exactly 500 per card though, as was claimed of Titans. It is pretty surprising, though.
Did TFIII have a nonreference design?

A GTX 780 at 1255MHz and 1.24V isn't a 1300MHz Titan at 1.3V either.

I forget where I saw the tests, but you'll see plenty of Titan owners claiming that their cards can pull 500W each, which isn't all that unreasonable if you extrapolate the power consumption from Thescyy's rig.


500w each is pretty unreasonable, since you've got a probably 160w+ cpu at least in there, if not more(wouldn't surprise me to see it over 200w) with that beefy overclock. I can see how it would get pretty extreme, though. I'm surprised a Kepler card could get that high without popping.

That is out of spec though, and definitely not reference design or typical use. This is a 290x overclocked within stock parameters, hence the should-be recommendation of a 650w PSU. I'm sure 290x could get even further down that road with some nice LN2 and bios mods.

Edit: In light of things, though, I could believe 400-450w per card for a more extreme overclock on water, maybe 500 or more on LN2.
---
Want that Shield!
Ball and Cup on ps mobile has framerate issues. -stargazer64
#26Xeeh_BitzPosted 10/26/2013 7:37:05 PM
Snuckie7 posted...
DarkZV2Beta posted...
Thescyy posted...
DarkZV2Beta posted...
Snuckie7 posted...
1300MHz runs at 1.3V? Yeah poke around on OCN a bit and you'll see. GK110 is a huge, power hungry chip. The only reason it doesn't consume much at stock is because it's clocked conservatively.


Got any links?


My computer would power off on a seasonic 850w bronze psu when I would bench with both my gpu's at 1255mhz+ along with my cpu at 4.8Ghz+. It would pull 1000w from the wall and power down about 5 seconds into the combined test every time.


That's not exactly 500 per card though, as was claimed of Titans. It is pretty surprising, though.
Did TFIII have a nonreference design?

A GTX 780 at 1255MHz and 1.24V isn't a 1300MHz Titan at 1.3V either.

I forget where I saw the tests, but you'll see plenty of Titan owners claiming that their cards can pull 500W each, which isn't all that unreasonable if you extrapolate the power consumption from Thescyy's rig.


I don't even use that much at 1300 @ 1.27v

Unless you think .03 volts is really gonna crank it
---
Best Windows ever, Windows 8.1!
#27DarkZV2Beta(Topic Creator)Posted 10/26/2013 7:37:57 PM
Xeeh_Bitz posted...
Snuckie7 posted...
DarkZV2Beta posted...
Thescyy posted...
DarkZV2Beta posted...


Got any links?


My computer would power off on a seasonic 850w bronze psu when I would bench with both my gpu's at 1255mhz+ along with my cpu at 4.8Ghz+. It would pull 1000w from the wall and power down about 5 seconds into the combined test every time.


That's not exactly 500 per card though, as was claimed of Titans. It is pretty surprising, though.
Did TFIII have a nonreference design?

A GTX 780 at 1255MHz and 1.24V isn't a 1300MHz Titan at 1.3V either.

I forget where I saw the tests, but you'll see plenty of Titan owners claiming that their cards can pull 500W each, which isn't all that unreasonable if you extrapolate the power consumption from Thescyy's rig.


I don't even use that much at 1300 @ 1.27v

Unless you think .03 volts is really gonna crank it


Well, every card is different.
---
Want that Shield!
Ball and Cup on ps mobile has framerate issues. -stargazer64
#28ThescyyPosted 10/26/2013 7:42:54 PM
[This message was deleted at the request of the original poster]
#29PanopictonguyPosted 10/27/2013 12:36:05 AM
Ok just to be sure, would it be safe to Cross fire this card at stock on a 750w PSU?

hypothetically speaking based on the info we have thus far?
---
Oh, you thought they made this game for you? You clearly bought the wrong game.
#30DarkZV2Beta(Topic Creator)Posted 10/27/2013 12:38:34 AM
Probably, at absolute stock.
---
Want that Shield!
Ball and Cup on ps mobile has framerate issues. -stargazer64