In which case I'd have to agree with bru as those figures aren't right at all for the pure board of a GTX780 with default power target.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
ASUS Radeon R9 290X DirectCU II Graphics Card Review
- Thread starter humbug
- Start date
In which case I'd have to agree with bru as those figures aren't right at all for the pure board of a GTX780 with default power target.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Where are those numbers actually from review wise, I don't recognise that watermark?
Having looked at a couple reviews, it's really very odd that Asus chose firstly such a quiet profile then such a loud profile, why nothing in between. The hardwarecannucks site suggest that the quiet profile is only marginally to slow to maintain clock speeds, it jumps such a large amount to the "high" profile which also makes it sound pretty loud, if they'd gone with something in the middle they would still have maintained higher clocks, no throttling but way less noise. then you have software for the guys who want to whack it to 100% for benching anyway, why not have a fan mode in the sweet spot, IE lowest fan speed possible with zero throttling?
Like I said though, haven't been impressed with Asus cards in donkeys years, nor asus boards now and never been impressed with their pricing.
I've had Asus mobos die, a Asus direct cu 5850 which wasn't great which died, got a Asus direct cu 7950 which blew, had all kinds of voltage locks and problems, while a good price I still sent it back and got a 7970 for £15 more in the end which wasn't voltage locked, louder but way fewer issues.
Having looked at a couple reviews, it's really very odd that Asus chose firstly such a quiet profile then such a loud profile, why nothing in between. The hardwarecannucks site suggest that the quiet profile is only marginally to slow to maintain clock speeds, it jumps such a large amount to the "high" profile which also makes it sound pretty loud, if they'd gone with something in the middle they would still have maintained higher clocks, no throttling but way less noise. then you have software for the guys who want to whack it to 100% for benching anyway, why not have a fan mode in the sweet spot, IE lowest fan speed possible with zero throttling?
Like I said though, haven't been impressed with Asus cards in donkeys years, nor asus boards now and never been impressed with their pricing.
I've had Asus mobos die, a Asus direct cu 5850 which wasn't great which died, got a Asus direct cu 7950 which blew, had all kinds of voltage locks and problems, while a good price I still sent it back and got a 7970 for £15 more in the end which wasn't voltage locked, louder but way fewer issues.
It's quite good Romanian tech site Reviewstudio and numbers looks legit, cause -
''... it came as a result of increasing the voltage from 1.25 to 1.4V.''
If we check TPU findings, numbers doesn't look unconventional.
''... it came as a result of increasing the voltage from 1.25 to 1.4V.''
If we check TPU findings, numbers doesn't look unconventional.
Last edited:
Be that as it may then Asus have set the TDP limit far to high or its not working as it should and the power draw just keeps climbing uncontrollably.
Asus are going to have a lot of RMA's from fried cards with peapole trying to find high stable overclocks by pushing the volts up and up while the power draw increase never stops until it goes bang.
250 Watts refrance to perhaps 400 and a largely reduced life span if your mad for overclocking.
I would be astonished if Asus had 600 Watts through those circuits 24/7 in mind, no way, it will kill the chip in minutes. its amazing it survived the bench.
I'd guess possibly moar PSU RMA's incoming.
A half decent PSU will disconnect the circuit or shut down completely before it becomes overloaded.
Sorry but that one really tickled me.
Ha... lol, that is funny. Google spelling checker..... honest guv.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
The numbers VERY much look unconventional, they total system power consumption on TPU is higher to start with and lower to finish. They need 1.25v to hit the same kind of clocks this Romanian site hit with 1.4v.
Increasing the voltage from 1.25 to 1.4v has a MUCH higher impact than raising voltage from 1.1 to 1.25v. That TPU image absolutely contradicts the Romanian's sites findings and directly says they are BS which is what they look like.
Romanian site says 380W stock 620W or something over volted, card only. TPU says SYSTEM goes from about 430W to 600W.
As power increases exponentially with voltage, 1.25-1.4 increases power a great deal more than 1.1-1.25v, as such I would expect a card needing 1.4V to increase in power more, but why on earth they would need 1.4v to hit a low overclock no other reviews show the card needing says that in itself is where the problem is. They've just whacked voltage up to max for no reason and seen how high it can clock, regardless of the fact that they could do that clock for much less.
Like I said, you could put a 900Mhz core to 1.5v and probably double wattage for no performance gain... it's just stupid.
Conventional 'kill-a-watt' power meters used by majority of reviewers have quite low polling frequency, so are showing only avg values. Whereas in ms timescale power consumption variance (let's say gauged by dso) is much more 'entertaining'.
Jeh, but you should note that TPU are NOT using Intel Core [email protected] as a testbed.
Jeh, but you should note that TPU are NOT using Intel Core [email protected] as a testbed.
Last edited:
I think there is probably more to it than that, we all know what happens once the GPU's power draw hits the set limit, it down clocks and drops the volts. this is why we add 20% in MSI AB. some use hacks to add 50%.
Whatever has gone on here it is not drawing 600 Watts under any normal circumstance, the board Power Line limit been tweaked to pull way way way more than intended.
You have to remember that those voltages on a Titan, had it pulling near 500w in my testing (~460W). I had it connected to a MM only and couldn't believe what I was seeing and that was at 1.375V. It does look silly high but not in the realms of impossible.
Is that the one you killed? what did you do to the power target?
That was on a 125 power target, which didn't seem to make any odds to the actual wattage that was being pulled. There are 300% BIOS's out as well.
Thats interesting, (not being funny or anything
) under normal circumstance it should limit the power drawn.Normal Circumstance being, it having the original unmolested BIOS and power limits set by a properly functioning application like MSI AB. IE providing its also working correctly.
See my picture.
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
http://www.hardocp.com/article/2013/12/23/asus_r9_290x_directcu_ii_oc_video_card_review/10#.UrnFF7TwyhY
Another review for the op. Pretty much matching a gtx780 ti in all games tested.
Another review for the op. Pretty much matching a gtx780 ti in all games tested.