GeForce GT 240: Low Power, High Performance, Sub-$100

[Computer_Lots]

This card is not the fastest one that doesn't require the PCI-E power connector. There are several versions of the 9800GT low power edition which do not require the 6-pin power. They are priced around $80 - $90 which make them cheaper than the GT240 and more powerful. I'd like to see a comparison between these 2 cards as for power consumption and also to see if the low power editions suffer in performance compared to their high power counterparts. Here are a few links from Newegg for the cards I'm referring to...

http://www.newegg.com/Product/Product.aspx?Item=N82E16814162032

http://www.newegg.com/Product/Product.aspx?Item=N82E16814127442

http://www.newegg.com/Product/Product.aspx?Item=N82E16814500113




http://www.newegg.com/Product/Product.aspx?Item=N82E16814143186

 

[Computer_Lots]

OK, The links didn't work. Let's try again...

http://www.newegg.com/Product/Product.aspx?Item=N82E16814162032

http://www.newegg.com/Product/Product.aspx?Item=N82E16814127442

http://www.newegg.com/Product/Product.aspx?Item=N82E16814500113

 

[Shadow703793]

[citation][nom]zulezule[/nom]http://store.steampowered.com/hwsurvey/videocard/Looking at what cards people actually have (8800gt mostly), I think there are very few that would want to upgrade to this. Give us something better, Nvidia! The only reason why Ati doesn't have a 90% market share right now is that they can't make 5800s and 5700s fast enough.[/citation]
And the 40nm yields are bad too.

 

[JofaMang]

The 40nm yields being bad are why they can't make them fast enough...

 

[Caffeinecarl]

[citation][nom]Ramar[/nom]I really can't justify this card when a Sparkle 9800GT is on newegg for the same price or less than these cards. Perhaps if energy costs are really important to you?[/citation]
I'll take a little less performance to save a lot of electricity. After a while, more efficient cards pay for themselves.

 

[namwons]

To me this article seems really biased towards Nvidia. As some have said earlier a better comparison would have been the HD 4770 (Toms should have have know this of course), same 40nm and GDDR5 and 10.1. Even though it has a 6pin, its still only rated at 80w vs 70w of the GT240. If they were compared the HD4770 would have definitely win as its comparable to 9800GT and GTS250 when overclocked.

 

[deadlockedworld]

Great article. I was really excited--until i went to check them out on newegg. Now im pissed. In the $90 to $115 price range? How stupid are they? Thats not even competitive with the 9600gt.

I would have bought this at $75-85 just because of the power usage... however it is not AT ALL competitive in the 9800gt/4770 price range.

 

[deadlockedworld]

LOL... i just read the other comments--and we all said the same thing.

 

[cinergy]

Why is it with Tom's and graphics cards that they always have the most über overclocked and custom passively cooled solutions for nvidia and for ATI its always the crappiest reference cards you can find? Oh, i forgot, its Tom's.

And no mention about HD4850 which is at the exact same price point as GT240? I tell you why, because HD4850 wipes the floor with GT240 and they don't want to show us that. GT240 is quite useless turkey compared to HD57xx lineup.

 

[roastmaster]

So where can you guy the palit GT240 1GB GDDR5???

 

[scrumworks]

[citation][nom]Uncle Meat[/nom]The memory on the Diamond Radeon HD 4670 is clocked 200Mhz below reference speeds.[/citation]

Well surprise surprise. Tom's playing fishy again with ATI and hoping no-one notices anything. I guess they spent some time to find the slowest HD4670 card out there.

 

[Guest]

The problem with this card is that both AMD and nVidia itself have cheaper cards that perform better. A Radeon 4770 or Geforce 9800 costs the same or less and kicks this cards ass all over the place.

 

[deadlockedworld]

its on newegg, for like $115

 

[scrumworks]

[citation][nom]rdhood[/nom]What dark_lord69 and noob2222 said. The 4670 is starting to see after-rebate prices of just $40. The $100 price point is closer to the 4770.[/citation]

Don't expect Cleeve to have any reasonable defense on that statement.

 

[Lavacon]

So why wasn't this compared to the 4850 again?

 

[cleeve]

[citation][nom]noob2222[/nom]Not a bad article, but very misleading however. Sub $100 tested with cards over $100? http://www.newegg.com/Product/Prod [...] -_-ProductAt that price its competing with the 4770s not the 4670. But this is an Nvidia article, gotta make them look good by omitting certain facts.[/citation]

Unfortunately, we were told it was solidly a sub-$100 offering and got to see the pricing when everyone else did, after the article went live.

Are we advocating paying $110 for a GT 240? Hells no. You're going to have to use your head a bit on that one. Get a 4850 instead. If it's $90, get an 8800 GT instead.

But will the prices settle down after launch? History tells us they'll quickly fall to where they should to be competitive. We're saying the DDR3 GT 240 is a good deal at 4670 prices, and the GDDR5 GT 240 is a good deal at just under 9600 GT prices.

If they stay too high, simply don't buy the card... not rocket science, just common sense.

 

[cleeve]

[citation][nom]scrumworks[/nom]Don't expect Cleeve to have any reasonable defense on that statement.[/citation]

See post above, scrummy.

 

[evolve60]

this just goes to show that the old 8/9 series from nvidia can still hold their own in some way, concidering that these cards were released almost if not more then 3 years ago, and still held its own against some of the 4k series from ATi, until recently when nvidia stopped releasing cards at the sub $100 market

 

[sublifer]

It also surprised us how little power the GeForce 9600 GT uses under load, as we expected a much higher power draw from that card.

I understand using the low power athlon x2 to test idle voltages but then when its time to test under load you're going to be cpu limited and would not be stressing the gpu enough to get an accurate load measurement. Maybe you should test load measurements with a nice beefy i7 or Phen x4 so you're actually putting a load on the gpu.

 

[cleeve]

[citation][nom]sublifer[/nom]Maybe you should test load measurements with a nice beefy i7 or Phen x4 so you're actually putting a load on the gpu.[/citation]

Using a CPU that draws more power would simply show us higher power usage fro the CPU. GPU stress testers like Furmark don't usually stress the CPU.

 

[sublifer]

[citation][nom]Cleeve[/nom]Using a CPU that draws more power would simply show us higher power usage fro the CPU. GPU stress testers like Furmark don't usually stress the CPU.[/citation]
If use the same cpu to test load across all the GPU's then its still even.

 

[juliom]

[citation][nom]noob2222[/nom]Not a bad article, but very misleading however. Sub $100 tested with cards over $100? http://www.newegg.com/Product/Prod [...] -_-ProductAt that price its competing with the 4770s not the 4670. But this is an Nvidia article, gotta make them look good by omitting certain facts.[/citation]

Amen, brother.

 

[juliom]

[citation][nom]caffeinecarl[/nom]I'll take a little less performance to save a lot of electricity. After a while, more efficient cards pay for themselves.[/citation]

No they don't. I think someone has been taking too much caffeine

 

[scrumworks]

[citation][nom]Cleeve[/nom]See post above, scrummy.[/citation]

"we were told it was solidly a sub-$100 offering".

http://www.cultofmac.com/wordpress/wp-content/uploads/2009/11/DoubleFacePalm.jpg

 

[cleeve]

Are you suggesting we are at fault for being told incorrect pricing?

Aw scrummy, you can do better than that.