-
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.
What is wrong with the non ref 290Xs ?
- Thread starter Kaapstad
- Start date
For gaming I always use stock clocks too.
Soldato
- Joined
- 2 Jan 2012
- Posts
- 12,410
- Location
- UK.
Other reviews show the custom Asus 290X clocking really well...
Soldato
- Joined
- 6 Feb 2010
- Posts
- 14,583
What memory did your 290x have? My Eplida can only managed 1400MHz max...any higher I'd get black screen.
On the subject of overclock, my 290x manage to do 1200MHz on the core clock on +131mV.
Last edited:
Soldato
- Joined
- 5 Sep 2011
- Posts
- 12,883
- Location
- Surrey
Kaap has ASUS which I believe is Elpida. Personally I don't think it's all that much to do with RAM type, Elpida has somehow gotten a bad name but they still clock fairly well.
Soldato
- Joined
- 6 Feb 2010
- Posts
- 14,583
So may be it's really down to the batch then? Like the later cards are not providing adequate voltage to the memory...
^^^^ put the same card in the hands of the likes of you, me or Matt and show them how its done.
Holly Crapdo your neighbours lights dim?
All in a good cause
http://forums.overclockers.co.uk/showthread.php?t=18566653
Aren't reviewers supposed to get cherries? Not heard of the site in the OP so maybe they just got really unlucky off the shelf... no, no, I can't believe it!
With the amount of volts you can chuck at these things it's just not possible to be stuck below 1100...
With the amount of volts you can chuck at these things it's just not possible to be stuck below 1100...
lol did you read the voltage they applied to get 1090?
+ Infinity.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
I think there are two things happening, idiot reviewers whacking it right up to 1.4v then getting hit but huge power/heat and that is killing overclocking and second. 1250Mhz on the core isn't always 1250Mhz on the core. As in the card may only go 50Mhz higher in terms of actual clock speed, but it might be staying at that clock a lot more often than the stock card.
Who knows really, because on a stock card you can whack the gpu speed and voltage up, and it can work it will just throttle loads doesn't necessarily mean you'd get a higher stable max speed, but as before you'd be near it more often.
It's probably a combination of all of them, maybe overclocking tools as well. One issue I find is the voltage fluctuates a hell of a lot as it changes gpu speed due to either power or temp, so maybe you set 1300Mhz at 1.4v, but at one point it drops to 1250mhz and the voltage fluctuates with the clock speed and that combination isn't stable. I've not played with forcing constant voltage or know if that option even works with 290 cards yet.
I'm not convinced of quite a few of the third party coolers though, the high end direct cu/lightnings/other things used to come with triple slot coolers and AFAIK too many of them have now condensed these into dual slot coolers. But with one slot being fans, essentially the difference between a dual and trip slot cooling is all but doubling the surface area of the heatsinks. Asus's particularly doesn't impress me, seems to be a pretty narrow gap between fins, dual slot, the heatsink clearly has to be much lower than it appears around the fans, so while the heatsink looks quite high at the ends it's likely, over the majority of the surface, half that height. It's also got the fins going front to back so two fans with not much area for air to move in. If the fins were top to bottom the air could flow out much more easily. The temps weren't bad at stock but with 1.4v that heatsink is not adequate IMHO.
they may have found the clock speed say was fine at 1250Mhz, but it throttled down to 1090Mhz largely due to their 1.4v, so they decided 1090Mhz was the "top" speed it could do, who knows, most reviewers(based on meeting many) are as daft as anyone else.
Who knows really, because on a stock card you can whack the gpu speed and voltage up, and it can work it will just throttle loads doesn't necessarily mean you'd get a higher stable max speed, but as before you'd be near it more often.
It's probably a combination of all of them, maybe overclocking tools as well. One issue I find is the voltage fluctuates a hell of a lot as it changes gpu speed due to either power or temp, so maybe you set 1300Mhz at 1.4v, but at one point it drops to 1250mhz and the voltage fluctuates with the clock speed and that combination isn't stable. I've not played with forcing constant voltage or know if that option even works with 290 cards yet.
I'm not convinced of quite a few of the third party coolers though, the high end direct cu/lightnings/other things used to come with triple slot coolers and AFAIK too many of them have now condensed these into dual slot coolers. But with one slot being fans, essentially the difference between a dual and trip slot cooling is all but doubling the surface area of the heatsinks. Asus's particularly doesn't impress me, seems to be a pretty narrow gap between fins, dual slot, the heatsink clearly has to be much lower than it appears around the fans, so while the heatsink looks quite high at the ends it's likely, over the majority of the surface, half that height. It's also got the fins going front to back so two fans with not much area for air to move in. If the fins were top to bottom the air could flow out much more easily. The temps weren't bad at stock but with 1.4v that heatsink is not adequate IMHO.
they may have found the clock speed say was fine at 1250Mhz, but it throttled down to 1090Mhz largely due to their 1.4v, so they decided 1090Mhz was the "top" speed it could do, who knows, most reviewers(based on meeting many) are as daft as anyone else.
Last edited:
I know the guy in person responsible for the test and the overclocking result. And I can also say that they don't get "cherry picked" cards at all. Also the review site like 100% of all danish hardware reviews sites are managed by volunteers ! it's all non profit an no full time employed staff behind it. Most is driven by sponserships and things paid for out of your own pocket (like test systems etc).
To the test itself. The Asus GPU Tweaker was/is chosen because that it's Asus own OC tool and is what per default comes with the card.
Also regarding the OC ! going above 1100MHz core would result in missing textures in some spots in games. Therefore it's been concluded that the max OC setting is 1090Mhz. The card would actually go further but graphical glithes would appear eventhough the card was rock solid. Voltages was increased to see if that would get rid of abnormalities - but no. When they get graphical issues it's not accepted as a legit OC.
Basically this Asus card was a dud !.
To the test itself. The Asus GPU Tweaker was/is chosen because that it's Asus own OC tool and is what per default comes with the card.
Also regarding the OC ! going above 1100MHz core would result in missing textures in some spots in games. Therefore it's been concluded that the max OC setting is 1090Mhz. The card would actually go further but graphical glithes would appear eventhough the card was rock solid. Voltages was increased to see if that would get rid of abnormalities - but no. When they get graphical issues it's not accepted as a legit OC.
Basically this Asus card was a dud !.