The Myths Of Graphics Card Performance: Debunked, Part 1

[Mussels]

vsync is a no no for me and fps games, I can not handle any bit of input lag when playing those game. Even my Sony LCD TV has too much. For most other games it's fine, like civ 5/dota2/terraria/ etc.. but I do notice some games react badly with vsync and the lag is HUGE but I have not come across any of those recently.

set it to game mode. i game on a sony 46" these days. even in that silly test i'm always under 250ms.

 

[Nordic]

I just tried the reaction tests. Approaching 40 and reflexes at on average 300-320. This explains a lot to me now about why I lose in BF4 gunfights, lol. Age is such a bitch.

Don't feel too bad. I am in my 20's and I was getting the same average on the visual one. 20/20 vision but never great at sports. Maybe its that I use a tv for a monitor.

 

[MT Alex]

is this an exercise in how to twist the facts to make any point you wish?

Welcome to the wonderful world of qubit.

 

[RCoon]

Welcome to the wonderful world of qubit.

Sick and tired seeing the same old people take cheap shots at qubit just because of his past with TPU. Grow up both of you. If you don't like what he has to say, you don't have to read it, and maybe take a second thought before you post pointlessly to essentially insult the guy.

 

[Vario]

Toms Hardware is such a crappy site. Much love for this site and hardforum, every time I go over there (TH) I remember how much better it is here.

Even anandtech gets too caught up theoretical nonsense and their reviews don't tell me if its a good product.

 

[RejZoR]

Nonsense. You buy what you can afford and overclock the hell out of it. It really doesn't matter what it is, it can be overclocked. Even my mice and router are overclocked. I expect no less from any other component. Overclocking is free and only requires your time. Not doing it is just plain stupid, because it simply works. Gains in Natural Selection 2 that i'm playing a lot lately is just insane. Overclocking the HD7950 from 1000/5000 to 1100/6200 made a significant gain of 20fps (from 84 to 105) in main menu where the game shows fps counter with ingame 3D rendered background. I haven't bothered checking how it is ingame, but on same servers with 32 players, it feels significantly smoother. So, overclocking clearly works. I just can't see myself lowering the settings, i just can't do that, because the game turns into something what developers didn't intend during design time and that makes it crap. Spending tiny bit more when purchasing the card for better cooler like the WindForce 3X also helps with silence if you tune the fan profile curve to your preference. It will be dead silent, yet still within reasonable specs. it's performing so well i see very little point in buying any of the R9 series cards. maybe R9-3xx and since even that will be just a refresh, i probably won't even bother. Unless they bring something really interesting, like the GCN was to the old VLIW and i'll buy it out of curiosity.

 

[MT Alex]

Sick and tired seeing the same old people take cheap shots at qubit just because of his past with TPU. Grow up both of you. If you don't like what he has to say, you don't have to read it, and maybe take a second thought before you post pointlessly to essentially insult the guy.

That wasn't a cheap shot, Love, but I'll keep that in mind. And I'm not sure I like being lumped into the group of "same old people."

 

[Mussels]

that does seem like a cheap shot against qubit, and kinda uncalled for. his news articles didnt mesh with TPU's style and that was that.


i think the article was flawed based on chosen hardware. the lower end the hardware, the greater the overclocking gains. getting high end hardware (and high wattage!) with known thermal throttling issues was not a good way to test overclocking.

 

[Steevo]

Part of being an open forum is being called out when information is wrong, or facts are slanted. It wasn't Qubit that wrote the article that I am aware of, and (almost) all of us agree the article is slanted and provides little if any real information, so it seems like a fluff piece written for the sake of showing that brand X is better than Y in this far dark corner of obscure testing that no one in their right mind ever does, comparing retail cards to one press release card.

When someone does it consistently they paint a target on themselves.

 

[eidairaman1]

By the way best gaming experience is still on a crt monitor lol.

yes that article is a flawed way of thinking and should be disregarded by even average joes (those who dont overclock,or even know about it or afraid to). Once i have my entire parts set ill be running with vsync enabled to see if any major performance loss is there.

 

[RCoon]

And I'm not sure I like being lumped into the group of "same old people."

I apologise for generalising, I was just a little mad about it is all, as there are other more senior members who seem to be making a bit of sport about it recently.

 

[Vario]

By the way best gaming experience is still on a crt monitor lol.

yes that article is a flawed way of thinking and should be disregarded by even average joes (those who dont overclock,or even know about it or afraid to). Once i have my entire parts set ill be running with vsync enabled to see if any major performance loss is there.

I never run Vsync, the diagonal cut partial frames don't bother me but the latency increase does. I have good vision at 20/15.

 

[ZenZimZaliben]

For those that don't want to read the article.. Here is a link to the tests. I do worse on the Audio then the Visual by a large margin. Lowest I got on visual was 176ms, and average was 194ms. Audio was 230ms.

Visual - http://www.humanbenchmark.com/tests/reactiontime/
Audio - http://cognitivefun.net/test/16

 

[vega22]

No. They clearly showed, that when you limit the noise to 40dB(A) , the only card from the bunch that could actually improve at those fan settings was the Titan

yes, as that is very much a real world restraint for everyone. it is in no way is an unrealistic limitation which is used to to make a point at all.

Sick and tired seeing the same old people take cheap shots at qubit just because of his past with TPU. Grow up both of you. If you don't like what he has to say, you don't have to read it, and maybe take a second thought before you post pointlessly to essentially insult the guy.

if i knew what the hell you are on about i might take offence at this post dude, idk :/

save derailing this pm me so i have a clue

 

[Vario]

I think he just link dumps toms hardware but i am not sure. This and the steam thread...?

 

[Steevo]

On my actual PC at home I managed 180's on the audio, and 250-280 on visual, on my work laptop with a wireless mouse I was 200-250 on visual.

 

[erocker]

227 visual
134.67 audio

This is fun!

 

[ZenZimZaliben]

227 visual
134.67 audio

This is fun!

It's the little things am I right.

I did the tests on my work computer which is a POS on a POS monitor. Will test at home.

 

[erocker]

Me too. My mouse came with this thing.. In 2007 lol. Works fine though.

 

[eidairaman1]

I never run Vsync, the diagonal cut partial frames don't bother me but the latency increase does. I have good vision at 20/15.

Ouch corrective lenses. 20/20 here