Information

Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!



Discussion

Discussion on Article:
Nvidia GeForce GTX 465 Review: Right for Rehabilitation

Started by: trek554 | Date 05/31/10 12:56:54 AM
Comments: 34 | Last Comment:  10/20/10 07:35:41 AM

Expand all threads | Collapse all threads

[1-13]

1. 
what is wrong with the gtx465 that you got? all other reviews have the power consumption, heat, and noise below that of the gtx470. heck all other reviews got better overclocks too.

there is something wrong with your gtx465 for it to consume that much power, make that much heat and even generate more noise than the gtx480.
0 0 [Posted by: trek554  | Date: 05/31/10 12:56:54 AM]
Reply
- collapse thread

 
The 465 has a load core voltage of 1.025v. What do you expect? And the power consumption of the 465 is right at the heels of the 470:
http://www.anandtech.com/...vidias-geforce-gtx-465/14
Edit:
Ok, I looked at their 470 power consumption chart at 196.6 Watts it's definitely incorrect. But that doesn't mean that the 465 power consumption is incorrect. I think some sort of error was made with the 470.

0 0 [Posted by: CPUGuy  | Date: 05/31/10 06:35:55 AM]
Reply
 
look at ALL other reviews besides Anandtech though. xbit is the ONLY one that has a louder, hotter and more power consuming gtx465 than the gtx470. also xbit has the ONLY gtx465 that cant oc very well. common sense should tell you that the something is probably wrong with those particular cards that both Anandtech and xbit got. maybe its an issue with some Zotac cards.
0 0 [Posted by: trek554  | Date: 05/31/10 11:43:53 AM]
Reply
 
Ok, I looked at their 470 power consumption chart at 196.6 Watts it's definitely incorrect.


They are correct, because we re-checked GeForce GTX 470 twice and got same numbers. Our methodology is far more accurate, than any of other sites using consumer grade watt-meters. They measure system total power consumption from AC outlet, we measure graphics card consumption only. But as soon as i'll return to Tallinn again, i shall re-check it again - for the sake of accuracy.
0 0 [Posted by: Vader@Xbit  | Date: 06/01/10 02:45:44 AM]
Reply
 
I then ask that you test another 470 and see if you get the same results. Your website is the only one that has such low power consumption numbers for the 470. What good is it to publish such low power consumption numbers if no one else is seeing what you are seeing?

Therefore, I do ask that you test another 470 (and perhaps another motherboard) to see if those numbers pan out. Because as of right now, you're the only ones publishing such numbers.

Although you methodology is different from others your power consumption figures for the 480 are in line with other reviews. So methodology alone isn't the problem here.

0 0 [Posted by: CPUGuy  | Date: 06/02/10 08:37:32 AM]
Reply
 
So some are complaining that the GTX465 numbers are too high; so xbitlabs needs to check another sample. Now you are complaining about the GTX 470. Not all cards are the same. There is variance between different samples in the population. It is not in readers benefit to get similarly performing samples on every review website. That will create a fake certainty that all products on the market are the same.
Further, blaming xbitlabs for the power consumption on the cards is ridiculous. It's not like they shop around for the card with the highest/lowest consumption before they review it. This site uses the most scientific way to measure the power consumption and I hope that comments like yours do not make them use more mainstream approach. They use modified motherboard and a research grade equipment to measure the power consumption, not a meter at the outlet.
0 0 [Posted by: jonup  | Date: 06/02/10 12:51:28 PM]
Reply
 
Their 480 power consumption figures are in line with other reviewers while the 470 is not. Vadar already posted that they will do a review on a different card to see what the power consumption is.
0 0 [Posted by: CPUGuy  | Date: 06/03/10 01:24:55 PM]
Reply
 
I then ask that you test another 470 and see if you get the same results.


That's exactly what we plan to do.
0 0 [Posted by: Vader@Xbit  | Date: 06/03/10 12:58:04 AM]
Reply
 
Good to know.
0 0 [Posted by: CPUGuy  | Date: 06/03/10 01:22:58 PM]
Reply
 
The authors mentioned the fact that the GPU must be defective based on the higher power consumption. Even though their sample is not representating the entire population, if two websites are having similarly behaving GPUs this means that there are at least a few of them out there. Getting a good one would be a hit and miss. As for the noise, GTX465 is just barely louder than the gtx 470. This is to be expected as they use same cooling and the 465 is consuming a little more energy. The fan probably spins just a little faster.
0 0 [Posted by: jonup  | Date: 05/31/10 01:02:14 PM]
Reply
 
I couldn't have said it any better. One thing that trek doesn't seem to grasp is that these look to be harvested GPUs that simply didn't make the cut for a 480 or 470. So mileage will vary. I hope he realizes that before buying one.

0 0 [Posted by: CPUGuy  | Date: 05/31/10 04:25:27 PM]
Reply

2. 
Now look at TDP ratios of each 5870 and 465 and make a conclusion about DX11 performance (Unigine extreme tesselation for example). Is Fermi still looking that good ? No it isn't. Who forced you to write such bull$&!7 ? Fermi's architecture is a pure failure no matter the API.
0 0 [Posted by: 7eki  | Date: 05/31/10 06:16:54 AM]
Reply

3. 
It's pitiful that you omitted DX11 computing (OpenCL or CUDA).
Who cares about a few frames more in modern games now?
0 0 [Posted by: Azazel  | Date: 05/31/10 06:36:32 AM]
Reply
- collapse thread

 
I honestly don't see the need to discuss it in any detail. It's a moot talking point.
0 0 [Posted by: CPUGuy  | Date: 05/31/10 06:52:50 AM]
Reply
 
If so, what is the difference of this card and hd5770? It has some technology upperhands as well.
0 0 [Posted by: mtcn77  | Date: 05/31/10 02:34:10 PM]
Reply
 
Price, power, heat and performance for starters. Those are factors that customers find important. And I think the authors here covered that well. The rest is simply moot.
0 0 [Posted by: CPUGuy  | Date: 05/31/10 04:27:53 PM]
Reply

4. 
Hello Alexey and Yaroslav,

Your pros & cons is not consistent.

In the summary You wrote in Highs: "Relatively low noise"
and You wrote in Lows: "High noise"

So make up Your mind guys! Is it noisy or not ?
0 0 [Posted by: dratewka  | Date: 05/31/10 08:51:31 AM]
Reply
- collapse thread

 
It was quite noisy in our testing environment, as we mentioned hot weather. Thanks for correction! We'll correct this ASAP.
0 0 [Posted by: Vader@Xbit  | Date: 06/01/10 02:49:06 AM]
Reply

5. 
So far as I am concerned this card is nothing but fail, more heat than performance and for those who live below the arctic circle it is a major issue in the long run. Perhaps in deep space ware it is minus 200 below zero in the shade it might fair well.
0 0 [Posted by: nforce4max  | Date: 05/31/10 12:27:57 PM]
Reply

6. 
The Wolfenstein engine is too simple for the top-end graphics cards like GeForce GTX 480 and 470 to show their best. It is the sheer amount of hardware resources which is necessary here.


Is anyone still taking this Nvidia ball-licking "review" site seriously?
0 0 [Posted by: AsianJetlag  | Date: 05/31/10 04:00:26 PM]
Reply
- collapse thread

 
AsianJetlag, the first sentence that you quoted is the opposite of being so "pro-Nvidia" since it actually points out that the Fermi cards lack the sheer TMU power for a simple game like Wolfenstein or many other older games (where GT200 cards shine).
0 0 [Posted by: Bo_Fox  | Date: 06/01/10 12:39:30 AM]
Reply

7. 
After reading a bunch of reviews and extrapolating the results I must admit that Nvidia has disappointed us again. This card turned to be much worse than a 5830, which I thought to be unlikely. This entire Nvidia's generation should be avoided.
0 0 [Posted by: FLA  | Date: 05/31/10 06:13:40 PM]
Reply
- collapse thread

 
Several recent reviews on the 465 that also cover the 5830 show the 465 to be overall faster. However, at $280 compared against 5830's launch price of $240, yeah I do consider the 465 to be "worse" after considering higher consumption, heat output and noise.
0 0 [Posted by: Bo_Fox  | Date: 06/01/10 12:37:19 AM]
Reply
 
I think at the very high end there is a room for GTX480 and GTX480SLI. Liquid cooling these mosters would take care of the heat and noise. People who can afford such setups can afford the necessary cooling. For those folks performence is more important than price/performance ratio.
0 0 [Posted by: jonup  | Date: 06/01/10 07:37:43 AM]
Reply
 
You won't make a PC quieter with a water cooling solution. As you will still add just as many if not more fans to your PC then with just the video card alone.
0 0 [Posted by: CPUGuy  | Date: 06/01/10 02:57:32 PM]
Reply
 
You might not make it silent but it will definately be quiter than the stock cooler(s) on the GTX480(SLI).
0 0 [Posted by: jonup  | Date: 06/02/10 03:34:22 AM]
Reply

8. 
As per the usual, Xbit Labs has always been pro ATI, so most of the so called "conclusions" are mainly just conjecture. Seriously comparing the 465 to the 5850, are you kidding me? The 465 has been brought in to take on the spot between the 5830 and 5850, which it does very well. It is amazing that the 465 trades blows with the 5850, and it totally kicks the 5830's butt. That is why the card is priced the way it is, between the two.

Now when Nvidia releases the GTS450 or whatever it will be called it will compete with the 5830. You bad mouth a card that has chosen to take a middle ground where ATI does not have any card and you try to compare it to the slot above it then criticize it for being $50 less and not able to beat the 5850 enough.

You guys really need to rerun the GTX470 tests with a diff 470 as you guys are the only site on the net that has the numbers the way you do for power consumption. Either you got a hell of a 470 or you fudged the numbers.
0 0 [Posted by: daseto  | Date: 06/01/10 09:20:11 PM]
Reply
- collapse thread

 
No, Xbitlabs has NEVER been pro ATI simply because of the fact that they compare ATI's QUALITY Adaptive AA to Nvidia's TR-MSAA (not TR-SSAA).

This alone is the outstanding, profound evidence. Wake up.
0 0 [Posted by: Bo_Fox  | Date: 06/05/10 09:20:12 PM]
Reply

9. 
NV must be really desperate to even show the light of day to this dog of a card
0 0 [Posted by: alpha0ne  | Date: 06/01/10 09:32:15 PM]
Reply

10. 
OMG what a POS. Far fewer shaders than the 470 but it uses more power.

Who thinks up this stuff?
0 0 [Posted by: Aristide1  | Date: 06/02/10 08:09:18 PM]
Reply
- collapse thread

 
Simple, someone who thinks you will buy it. That's who.
0 0 [Posted by: CPUGuy  | Date: 06/03/10 01:22:32 PM]
Reply

11. 
Re-checked reference GeForce GTX 470 3D power consumption 5 minutes ago. The final result is 202.3 Watts under Crysis Warhead. Anyway, it is much lower, than GeForce GTX 465's 224.2 Watts. Sorry, guys, things are what they are.
0 0 [Posted by: Vader@Xbit  | Date: 06/07/10 09:15:30 AM]
Reply

12. 
all these solutions are good and attractive
0 0 [Posted by: mike1101  | Date: 06/16/10 03:05:00 AM]
Reply

13. 
It's a very interesting and in depth article.

I picked up a Zotac GTX 465 for £150, but you can get the cheapest models for around £145 at the time of writing in the UK. The cheapest Ati Radeon HD 5850 is around £182 which is a considerably higher price while the Ati Radeon HD 5870 costs of lot more with the cheapest around £260. The Zotac card also comes with the added bonus of a long extended warranty of 5 years which is longer than anyone is likely to want to keep any graphics card anyway because of it becoming obsolete.

Yes the Zotac GTX 465 may perform with slower frame rates than the Ati 5850 on the majority of tests, but it's considerably less slower at 1600x900 resolutions which is all I ever need to use anyway and it's only slightly slower while overclocked, although on around 20-25% of games it's actually faster which in my opinion isn't bad for a cheaper card. One would expect considerably lower frame rates than the Ati Radeon HD 5870 which is a lot more expensive, similarly to the more expensive Nvidia GTX 470, but the Nvidia GTX 480 is around twice the price and is therefore in a different league. One also needs to consider that all modern Nvidia cards including the GTX 465 come with the extra hardware required to run the powerful Nvidia PhysX engine which is something Ati cards simply don't have. When the PhysX engine is used to it's full potential it can be quite a noticeable advantage to have, take a look at comparisons of some games with PhysX enabled against PhsyX not enabled on YouTube and you'll see the difference. More and more games are starting to use the Nvidia PhysX engine and while some only offer small improvements, others offer much better effects and true realism with the PhysX engine enabled.

In the article the Zotac GTX 465 graphic card was directly compared to the older Nvidia GTX 275 on various games, the Zotac card performed better on most occasions, although not usually by much and occasionally it was even slightly slower. Well one should also consider that the Nvidia GTX 275 has held it's price well and the cheapest model still costs a little more than the Nvidia GTX 465 at the time of writing in the UK which to me makes the Nvidia GTX 275 a very poor buy in my opinion, especially since the Nvidia 200 series of graphics cards don't support enhanced DirectX 11 graphics and pixel shader 5.0. What's more, some games where tested using enhanced Pixel Shader 5.0 graphics on the Zotac GTX 465 against inferior Pixel Shader 4.0 graphics on the Nvidia GTX 275, but usually if you reduce the graphics to Pixel shader 4.0 on a card that's capable of pixel shader 5.0 (often this can be achieved by running it in Windows XP or sometimes in the graphics options), it increases the frame rate noticeably since it's not doing as much work, so comparing the frame rates side by side isn't really fair in this case. With this in mind, in my opinion the Zotac GTX 465 or in fact any Nvidia GTX 465 series card is much better than the older Nvidia GTX 275 which surprisingly still costs slightly more in the UK at the time of writing.

My Zotax GTX 465 card overclocks better than in the article and is still running quite cool, this is probably because I have set-up a more aggressive custom fan profile which can be a little noisy when the card is pushed hard, but it's in my opinion worth it for improved performance. The card is a little hungry on power, but it's still much better than older graphics cards, I used to own a 512MB Ati Radeon HD 2900XT which was the fastest card ATI sold only a few years ago. It was also one of, if not the 1st card to support DirectX 10.1 and pixel shader 4 in it's day and it even officially supported HD graphics which was quite new then. The Ati Radeon HD 2900XT has a similar power consumption to the GTX 465 which was enormous in it's day when a 550W power supply unit was considered high spec and was the barest minimum requirement for running this card. This Ati Radeon HD 2900XT card also got extremely hot and was very noisy indeed with barely any room left for overclocking. It's performance was absolutely incredible back then and it still amazingly plays all the latest games today, some even on high settings, but it's performance still doesn't come close to the GTX 465 despite using a similar amount of power. In fact even today the Ati Radeon HD 2900XT compares to cheap modern entry level 3D graphics cards that cost around £50-£60 like the ATI Radeon HD 5570 which performs at similar frame rates with the addition of having DirectX11 support.

In conclusion, the Ati Radeon HD 5850 is a better card for higher frame rates on around 80% of games and the Nvidia GTX 465 struggles a lot more at very high resolutions. The Ati Radeon HD 5850 also uses less power, but you certainly lose out on the Nvidia PhysX engine and it also costs a bit more in the UK at the time of writing while the Nvidia GTX 465 is the cheapest of the cards compared in the article. I wouldn't recommend an Nvidia GTX 465 if you plan on running at higher resolutions than 1600x900 as the performance is then considerably less than the HD 5850, instead I'd either buy an HD 5850 or above, or save up for the more expensive Nvidia GTX 470 or even better the best, but the most expensive Nvidia GTX 480 as you then get the best of all worlds with the PhysX engine and good performance at super high resolutions. The Nvidia GTX 275 is now old, expensive at the time of writing and doesn't support DirectX 11 / Pixel Shader 5, plus it's usually performs slower than the Nvidia GTX 465, especially when comparing 2 games that only use DirectX 10.1 / Pixel Shader 4 or below.
0 0 [Posted by: pjcnet15  | Date: 10/20/10 07:35:41 AM]
Reply

[1-13]

Back to the Article

Add your Comment