The Myths Of Graphics Card Performance: Debunked (Toms Hardware)

[MentalIlness]

A month old article, but informative nonetheless.

http://www.tomshardware.com/reviews/graphics-card-myths,3694.html

 

[BadThad]

Great article, thanks!

 

[waggy]

blah. the V-card forum here is biased. But [redacted] tom's puts it to shame. or they used to years ago. i stopped going because it was so slanted it was insane. it's one of the last places i would go to for anything.

I used to love the place too. but [redacted] it got insane

Profanity isn't allowed in the technical forums.
-- stahlhart

 

[Ken g6]

Hm, they claim that small amounts of memory use over 2GB, e.g. 2093MB, mean that you need a 3GB card for that scenario. But I thought video cards would swap out some of their memory to main memory in that case. And if what they swap out isn't frequently used, won't a 2GB card suffice in that case?

 

[norseamd]

used to read toms since like 2010 when i had to research video cards after my 9800 gt blew up. back then it seemed like a good place to read stuff but gradually it seems to have worse and worse articles.

 

[VulgarDisplay]

I don't understand the point of their overclocking test.

 

[wand3r3r]

You can count on toms to test a thermally limited 290x by sliding the sliders over and verifying that it didn't do anything.

 

[TreVader]

Toms has been garbage since 2005, prob long before that but that's when I started building computers.

 

[flexy]

What exactly is this "myth"?

All I see is a table with a few cards listed and their FPS.

 

[BrightCandle]

Its a useful test. At any given volume of fan noise what is the resulting performance. Its a proxy somewhat for performance/watt testing but focussed more on what the user cares more about which is noise. If 40Db is what you want (I personally have 35Db so that is a bit loud for me) then this tells you which card performs the best at that level.

Its quite interesting for example to see a 290 and 690 both fail to provide any benefits of overclocking due to thermal limitations whereas the Titan still has headroom at this low volume setting.

Toms was trying to make a point about the trade offs between power/noise and performance, price etc and how AMDs cards where targeting a different audience to the 780/Titan's.

 

[taltamir]

it is odd that they mention G-Sync and FreeSync as future solutions that take out the need to choose situationally whether to enable or disable vsync... Yet for some reason they don't acknowledge the existence of adaptive vsync

 

[TreVader]

It's funny how tech sites look at each other. Back in 03/04 I was a member of B3D forums and I remember this forum was regarded as only a bit better than toms. [H] was regarded as better than AT but crude and vulgar.

 

[norseamd]

It's funny how tech sites look at each other. Back in 03/04 I was a member of B3D forums and I remember this forum was regarded as only a bit better than toms. [H] was regarded as better than AT but crude and vulgar.

any other forums you have general info on?

crude and vulgar is fine but empathy is appreciated

 

[bystander36]

it is odd that they mention G-Sync and FreeSync as future solutions that take out the need to choose situationally whether to enable or disable vsync... Yet for some reason they don't acknowledge the existence of adaptive vsync

Adaptive V-sync does not take away the need for G-sync and FreeSync, as it still allows tearing when you fail to reach your refresh rate. It is still a compromise.

 

[bystander36]

Its a useful test. At any given volume of fan noise what is the resulting performance. Its a proxy somewhat for performance/watt testing but focussed more on what the user cares more about which is noise. If 40Db is what you want (I personally have 35Db so that is a bit loud for me) then this tells you which card performs the best at that level.

Its quite interesting for example to see a 290 and 690 both fail to provide any benefits of overclocking due to thermal limitations whereas the Titan still has headroom at this low volume setting.

Toms was trying to make a point about the trade offs between power/noise and performance, price etc and how AMDs cards where targeting a different audience to the 780/Titan's.

This is what I took away from it. I personally care about noise levels. It does seem that a lot of the vocal crowd doesn't, but would bet a lot of people care about noise levels when they game.

They also talked about a lot of things other than OCing, and noise.

The only thing I disagree with the article was when talking about latency. They failed to mention how low latency effects how well you can track with your mouse and focused entirely on how quickly you can press a button, which have very different levels of sensitivity.

 

[MagnusTheBrewer]

Tom's is hardly an unbiased tech site. If it isn't Intel or Nvidia, it gets slammed in the articles and put down in the forums. They tolerate very little criticism of their methods, conclusions or, promotion of any other products. They've been very consistant though, their biases haven't changed in 15 years or longer.

 

[f1sherman]

any other forums you have general info on?

crude and vulgar is fine but empathy is appreciated

cross-forums talk is like stopping a random woman on the street and complaining to her about your wife

 

[Mand]

Tom's is hardly an unbiased tech site. If it isn't Intel or Nvidia, it gets slammed in the articles and put down in the forums. They tolerate very little criticism of their methods, conclusions or, promotion of any other products. They've been very consistant though, their biases haven't changed in 15 years or longer.

That must be why they gave the 290x the Elite Buy award, eh?

 

[taltamir]

Adaptive V-sync does not take away the need for G-sync and FreeSync, as it still allows tearing when you fail to reach your refresh rate. It is still a compromise.

1. I didn't say it takes away the need for g-sync/freesync
2. How is this justification for pretending it doesn't exist?
3. No, it isn't a compromise, it is just imperfect. Tearing is the default state not something you introduced by the feature. Having vysnc set to enable is a compromise (eliminate tearing, lose FPS). Adaptive vysnc obsoletes the "don't use vsync" choice such that the only 2 legitimate choices are "vsync-on" (the compromise feature) and "vsync-adaptive" (the imperfect non compromise)

 

[Makaveli]

Toms has been garbage since 2005, prob long before that but that's when I started building computers.

I c you've been around aslong as me.

I use to read toms all the way back from 2000-2005 and would post regularly. The some people left and it slowly went downhill. I once in a blue moon look at some of the articles posted but don't bother with the forums since then.

It really did get pretty ridiculous.

 

[MagnusTheBrewer]

That must be why they gave the 290x the Elite Buy award, eh?

It's mighty hard to fudge the same tests everyone else is doing. Of course, that never stopped them before.

 

[Silverforce11]

Performance normalized to noise with reference cards will always make AMD look bad (for several generations now!) and they know it and they DGAF per their interviews. They make reference cards to meet OEM needs only. If people don't like the noise, they can buy the custom variants from their AIB partners.

 

[Enigmoid]

Tom's is hardly an unbiased tech site. If it isn't Intel or Nvidia, it gets slammed in the articles and put down in the forums. They tolerate very little criticism of their methods, conclusions or, promotion of any other products. They've been very consistant though, their biases haven't changed in 15 years or longer.

Are you kidding?

They gave the MSI GX 60 a 'smart buy award'.

Toms forums are much more budget orientated and the forums are full of generally anti intel sentiment (its much harder to find anyone advocating a i3 over a 6300 for a build as opposed to these forums). If anything I would say the recommendations on AT are much more intel orientated likely due to some extent that people here tend to have larger budgets.

Toms forums are much less moderated.

AT seems to have an apple bias. They also have "AMD center".

 

[bononos]

.......

AT seems to have an apple bias. They also have "AMD center".

Did AT explain how the AMD center came about? I mean, AT used to (not sure about now) claim that it does not rely on advertisements etc.

 

[bystander36]

Tom's is hardly an unbiased tech site. If it isn't Intel or Nvidia, it gets slammed in the articles and put down in the forums. They tolerate very little criticism of their methods, conclusions or, promotion of any other products. They've been very consistant though, their biases haven't changed in 15 years or longer.

Here is the thing, you want to hear all the great things about AMD, and not the faults. Tom's does not ignore the faults. For people who care about those faults, it is a great site to read.

And what is crazy, is for all those who think the site is Nvidia/Intel biased, there have been many times people say just the opposite, which to me means they are unbiased.

If you want to ignore problems about a card, and skip ahead to strictly performance levels, there are other sites for you, but there are plenty of people who want to see the faults as well.