Spin Off: AT's Testing Methods & Uber Mode

[Saylick]

Our Ferrari F12 Berlinetta Review will only be made using the Wet Setup. Because turning the nob to the Race setup will rev the engine up to 10k rpm producing high decibels and we dont like to drive the car being that loud.
It is the first Ferrari we will not recommend because of high decibels produced by the 6262cc V12 engine.
/snip
Sorry if that was OT

Hahaha, interesting example but I don't think it's entirely fair to compare graphics card noise to the sweet, sweet sound of a race-bred V12.
That is, unless you want to be the "Harley Davidson of graphics cards".

 

[wand3r3r]

Well, what if Ryan said that while the 290 gives great gaming fps performance, the overall gaming experience was sullied due to the noise of the fan...then went into an objective explanation as to why the fan revved up as high as it did?

An opinion is an opinion, no matter how you may want to spin it in favor of said argument. If you don't like opinions here, then don't like opinions anywhere. And, for the fact of the matter...I don't have a problem with [H] given their opinion on the BF4 Open Beta performance between the two cards. You pretty much confirmed, whether knowingly or unknowingly, why opinions matter and why they are given and why they are not dismissed when they come from reputable review sites.

Yes, an opinion is an opinion, no arguments there. I can see the point you are bringing, however I don't agree that the subjectivity of the playability (while full of disclaimers and calls for more information) is on or even near the same level as the discrepancies on the 780 ti article brought up here. Of course the opinions of the reviewers matters, and it brought a lot of attention on the fan, however there are a number of people that can sacrifice noise for performance. The discrepancies are what are concerning to say the least.

 

[dust]

Compare this Ferrari to Nissan GTR.
The reviewer refuses to push GTR Launch button because it will be faster than Ferrari.

Rightfully so! Because the button wasn't pushed from the factory. That means it's not default setting. Jeez:sneaky:

 

[Z15CAM]

Rightfully so! Because the button wasn't pushed from the factory. That means it's not default setting. Jeez:sneaky:

ROLF

 

[Erenhardt]

Rightfully so! Because the button wasn't pushed from the factory. That means it's not default setting. Jeez:sneaky:

Hard to drive it thou, as it comes with 1-st gear on. Rest of gears are not factory default so we will see which car is fastest at gear it comes from factory.

The winner is GTR*!
Speed table
GTR : 35MPH (1st gear)
Ferrari: -16MHP (reverse)
*Testing concluded at factory default settings

Conclusion:
GTR was ininite times faster than ferrari

 

[Saylick]

Hard to drive it thou, as it comes with 1-st gear on. Rest of gears are not factory default so we will see which car is fastest at gear it comes from factory.

The winner is GTR*!
Speed table
GTR : 35MPH (1st gear)
Ferrari: -16MHP (reverse)
*Testing concluded at factory default settings

Conclusion:
GTR was ininite times faster than ferrari

I thought the car comes out of the factory with its engine OFF? Don't see how numbers other than zero could be possible. :whiste:

 

[TerryMathews]

I thought the car comes out of the factory with its engine OFF? Don't see how numbers other than zero could be possible. :whiste:

Well, ya know, they do call it an ignition switch...

 

[Erenhardt]

Well, ya know, they do call it an ignition switch...

...and it dramatically changes performance, heat and noise...

 

[Atreidin]

Well, ya know, they do call it an ignition switch...

OK but who would go out of their way to use it? It is simply absurd to think somebody who paid money for a product would try to understand how to properly operate it. What are we, a bunch of Einsteins?

 

[AtenRa]

Hahaha, interesting example but I don't think it's entirely fair to compare graphics card noise to the sweet, sweet sound of a race-bred V12.
That is, unless you want to be the "Harley Davidson of graphics cards".

heh, actually i still have a NVIDIA FX5800 Ultra from Gainward.

 

[AtenRa]

Seriously now, i do care about high noise video cards, but at the high-end we first care about Performance, then price and noise is the last thing we take in to consideration.

 

[Majcric]

In a Nutshell, I think this thread is begging for a showdown between AMD's 290X uber vs the GTX 780ti uber. Cranks the fans, voltage, whatever, then declare a winner.

 

[Janooo]

In a Nutshell, I think this thread is begging for a showdown between AMD's 290X uber vs the GTX 780ti uber. Cranks the fans, voltage, whatever, then declare a winner.

+1
... and an apology for the deception.

 

[Abwx]

however there are a number of people that can sacrifice noise for performance. The discrepancies are what are concerning to say the least.

Not much or even notat all , actualy it looks like an Nvidia card is quieter in isolation but once mounted on an actual case it will require the case fans to spin faster than AMD cards since Nvidia made the choice to blow hot air inside the case since it allow their cards to run cooler but at the expense of the ambiant inside the case.

 

[Teizo]

Not much or even notat all , actualy it looks like an Nvidia card is quieter in isolation but once mounted on an actual case it will require the case fans to spin faster than AMD cards since Nvidia made the choice to blow hot air inside the case since it allow their cards to run cooler but at the expense of the ambiant inside the case.

This is not true unless you have a crappy case with crappy fans. My SLI set up is dead quiet in idle and very quiet under load. So much to the point I look over to make sure it is running even though I know it is.

All you have to do is chose quality fans. As in all things...you get what you pay for. And to your point on that...using that argument...the 290 and 290X will make things even worse once the AIB models come out with their open airdesigns. However, even with their higher temp target, as long as someone has good airflow and good fans it shouldn't make too much difference...though there is a definite difference between dumping 80C into a case vs 95C into a case.

 

[Erenhardt]

This is not true unless you have a crappy case with crappy fans. My SLI set up is dead quiet in idle and very quiet under load. So much to the point I look over to make sure it is running even though I know it is.

All you have to do is chose quality fans. As in all things...you get what you pay for. And to your point on that...using that argument...the 290 and 290X will make things even worse once the AIB models come out with their open airdesigns. However, even with their higher temp target, as long as someone has good airflow and good fans it shouldn't make too much difference...though there is a definite difference between dumping 80C into a case vs 95C into a case.

Yes, more convection when the air is hotter, and that would be about it.
What now? We will go from not including factory settings, to testing each possible power/fan/clock setup?

 

[3DVagabond]

This is not true unless you have a crappy case with crappy fans. My SLI set up is dead quiet in idle and very quiet under load. So much to the point I look over to make sure it is running even though I know it is.

All you have to do is chose quality fans. As in all things...you get what you pay for. And to your point on that...using that argument...the 290 and 290X will make things even worse once the AIB models come out with their open airdesigns. However, even with their higher temp target, as long as someone has good airflow and good fans it shouldn't make too much difference...though there is a definite difference between dumping 80C into a case vs 95C into a case.

You aren't running 780ti in SLI. They'll produce more heat and noise.

I see people referring to the temp of the chip as meaning anything about heat produced, but it doesn't. It's the amount of energy dissipated that matters. Just as an example, if both cards use 200W they will both produce the same amount of heat. It doesn't matter if one is running at 95° and the other is running at 80°.

 

[Silverforce11]

This is not true unless you have a crappy case with crappy fans. My SLI set up is dead quiet in idle and very quiet under load. So much to the point I look over to make sure it is running even though I know it is.

All you have to do is chose quality fans. As in all things...you get what you pay for. And to your point on that...using that argument...the 290 and 290X will make things even worse once the AIB models come out with their open airdesigns. However, even with their higher temp target, as long as someone has good airflow and good fans it shouldn't make too much difference...though there is a definite difference between dumping 80C into a case vs 95C into a case.

We've already have multiple sites show a large drop in temps, noise AND power use for the R290/X on custom open air coolers. Its around a 40W drop in power use when it runs at lower temps and not 95C.

Pretty significant.

 

[Attic]

I tested a friends 290x. He had it in quiet mode despite being a strong computer/game enthusiast. This gave me a better appreciation for the concerns/stance/policy Anandtech has cited as regarding explanations for recent evaluations of the 290x vs competition. If I had not gone over I doubt very much that the switch on the card would have been used. He just didn't understand it despite having concerns for frames in BF4.

Subjectively, and I'm a guy who values silence, the card wasn't loud in uber mode at fan speeds up to 50%. The card didn't stand out vs other fans in a well ventilated case and the case itself was not loud. We never got it higher than 50% fan speed even when overclocking, card got to 93C. Bigger concern for me was some image corruption in Heaven 4.0 on the copper roofs that I noticed on his card at stock settings. I know I don't see the same image issues I saw on his setup as I do on mine, but I didn't iron it down to definitely being his card vs other variables. The image corruption was solid colored (red,white,other) small squares showing up in the copper roofs texture in one scene.

 

[Erenhardt]

I tested a friends 290x. He had it in quiet mode despite being a strong computer/game enthusiast. This gave me a better appreciation for the concerns/stance/policy Anandtech has cited as regarding explanations for recent evaluations of the 290x vs competition. If I had not gone over I doubt very much that the switch on the card would have been used. He just didn't understand it despite having concerns for frames in BF4.

If anytyhing, that is the reason why reviewers shouldn't discard uber profile. They should bring viewer's attention to double bios switch. Show the features, so less "enthusiast" will miss them, or be confused about them.

 

[Will Robinson]

In a Nutshell, I think this thread is begging for a showdown between AMD's 290X uber vs the GTX 780ti uber. Cranks the fans, voltage, whatever, then declare a winner.

There's really no need.
R290X is priced to compete with the $500 GTX780 which it easily defeats.
GTX780 Ti is a $700 card and it would be a complete failure if it didn't beat the R9 290X at $550

 

[wand3r3r]

This is not true unless you have a crappy case with crappy fans. My SLI set up is dead quiet in idle and very quiet under load. So much to the point I look over to make sure it is running even though I know it is.

All you have to do is chose quality fans. As in all things...you get what you pay for. And to your point on that...using that argument...the 290 and 290X will make things even worse once the AIB models come out with their open airdesigns. However, even with their higher temp target, as long as someone has good airflow and good fans it shouldn't make too much difference...though there is a definite difference between dumping 80C into a case vs 95C into a case.

What does aftermarket weak mid range cards have to do with it (strawman)? It does nothing to refute the point.

 

[96Firebird]

If anytyhing, that is the reason why reviewers shouldn't discard uber profile. They should bring viewer's attention to double bios switch. Show the features, so less "enthusiast" will miss them, or be confused about them.

Moving on, AMD’s dual BIOS functionality is back once again for the 290X, and this time it has a very explicit purpose. The 290X will ship with two BIOSes, a “quiet” bios and an “uber” BIOS, selectable with the card’s BIOS switch. The difference between the two BIOSes is that the quiet BIOS ships with a maximum fan speed of 40%, while the uber BIOS ships with a maximum fan speed of 50%. The quiet BIOS is the default BIOS for the 290X, and based on our testing will hold the noise levels of the card equal to or less than those of the reference 7970.

http://www.anandtech.com/show/7457/the-radeon-r9-290x-review/6

It's in the 290X review, what more do you want?

 

[Erenhardt]

http://www.anandtech.com/show/7457/the-radeon-r9-290x-review/6

It's in the 290X review, what more do you want?

I want both settings is all reviews, not introductory one only. It is a lot to ask people to check testing methods each time they look at graphs, to see if that is quiet mode, uber mode, or noise equalized mode.

 

[KingFatty]

When it comes to the problem of multiple modes/fan profiles, I think we could approach it sort of like a car review - tell us the Max horsepower and torque, and we'll understand what that means and extrapolate for all the various gears/transmissions/etc. because there are infinite range of RPMs you can run a motor so it's a waste to test all gears at all RPMs.

So for video cards, I read to review to know what the card is capable of doing as its max performance, whether it's enough horsepower for me to be happy. I don't need to know all the details of all kinds of settings/fan profiles, but I want to know the max capability. Sort of like knowing the capability of the silicon. I need to know that because I might watercool or use an aftermarket cooler, so telling me about the capability of the silicon is useful.

Sorry for the car analogy, but I think it's good to think like that where even a car with higher peak horsepower can still lose a race with another car, so if all you are doing is "racing" the cards against each other than that's not enough. We want to know enough about the card to be able to determine if we want to buy it.