[PCPER] Visual Effect of Vsync test yourself

[Imouto]

Maybe around the time that $300 2560x1440 monitors were discovered? I really don't know. I would love to have a single gpu card to do that resolution as well as my sli setup but outside of a titan for $1000 it doesn't exist.

I just wanted to know some numbers and Steam survey would throw some light. If you're about 1% of ppl out there I really don't know why this is blown so much out of proportion.

It's like crying frigging rivers when a Ferrari is doing 300 kph, a Lambo is doing 320 and 99% of ppl is driving a Hyundai.

7 articles about the matter in 3 months is being obnoxious.

 

[VulgarDisplay]

This is the fourth paragraph:



The emphasis is from the original. Seems pretty clear to me.

His emphasis on SLi titans providing a completely smooth and runt free experience, and then going on to say that AMD's engineers lied to him prove show bias.

All he had to do was actually TRY crossfiring two 7970's and comparing the results to the Titan SLi with Vsync for the purpose of this article. Instead he gave a completely lopsided test scenario and then went on to call AMD liars and then praise the Titan SLi Vsync experience as being smooth which blows my mind. How can people not interpret the articles this way?

This article is a joke, as all of his recent work has been.

Regardless of your hardware vendor preference if anyone can read his article and come to these conclusions it should tell him he's doing something wrong. I'm not the only person who reads the articles in this light. He should not have disclosed any of the hardware used in this test scenario because it was 100% not necessary, and he also could have done without mention AMD or Nvidia whatsoever.

 

[bystander36]

His emphasis on SLi titans providing a completely smooth and runt free experience, and then going on to say that AMD's engineers lied to him prove show bias.

All he had to do was actually TRY crossfiring two 7970's and comparing the results to the Titan SLi with Vsync for the purpose of this article. Instead he gave a completely lopsided test scenario and then went on to call AMD liars and then praise the Titan SLi Vsync experience as being smooth which blows my mind. How can people not interpret the articles this way?

This article is a joke, as all of his recent work has been.

Regardless of your hardware vendor preference if anyone can read his article and come to these conclusions it should tell him he's doing something wrong. I'm not the only person who reads the articles in this light. He should not have disclosed any of the hardware used in this test scenario because it was 100% not necessary, and he also could have done without mention AMD or Nvidia whatsoever.

Are you just trying to ignore what is written? HE IS NOT COMPARING CARDS! HE IS NOT SAYING CROSSFIRE OR AMD CARDS DO NOT PERFORM WELL! HE IS NOT COMPARING AMD TO NVIDIA, 7970 TO TITAN OR ANYTHING RELATED TO BRAND!!!

He made it VERY clear, he is comparing v-sync at a constant 60 FPS, and v-sync when it fails to reach a constant 60 FPS. Is that so difficult to understand?

 

[Rikard]

He should not show the 50% and 20% in the blind test, that defeats the purpose. We are comparing 60 and 30 FPS after all, not 12 and 6 FPS!

At a 100% speed I could tell them apart but once I had to rewind and see again. Vsync is much, much better than constant 30 FPS though... It is nice the he wrote he wants to take a look at input lag next. I have been hoping more reviewers would take that seriously.

 

[willomz]

I just wanted to know some numbers and Steam survey would throw some light. If you're about 1% of ppl out there I really don't know why this is blown so much out of proportion.

1% of steam users sure, but actually a much higher percentage of people who regularly read hardware news or say contribute to hardware forums.

Also just because you don't own an SLI/CF solution doesn't mean you haven't considered it as an option and want to know the pros and cons.

EDIT: I mean only 0.0001% of people would even consider the Corsair 900D, so why would Anandtech even review it?

 

[Imouto]

EDIT: I mean only 0.0001% of people would even consider the Corsair 900D, so why would Anandtech even review it?

Would Anand review it 7 times over 3 months? Don't frigging think so.

Look, some finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.

 

[Xarick]

I only watch these to listen to the music.

 

[cmdrdredd]

Would Anand review it 7 times over 3 months? Don't frigging think so.

Look, some finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.

To illustrate the changes due to driver revisions. different games, new software with frame limiters and new methods of finding frame times etc.

 

[bystander36]

Would Anand review it 7 times over 3 months? Don't frigging think so.

Look, some finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.
Look, same finds.

They are the first people to analyze SLI and CF in this manner. No one else has barely done anything. Every time they give a review of it, tons of people ask for them to test other conditions. Most of the requests come from AMD users who want to see if their configuration suffers these problems, or looking at ways to disprove what they showed. Rather than ignoring those people, they continue to dig into all those aspects.

They have several new things to explore. Latency, v-sync, and different card setups, like the 7990, when it comes out. Expect there to be a few others. Considering they are the only ones digging deep into this type of study and that it has never been done, people are interested. I know I am.

They have also only done 5 different SLI vs CF comparisons with different cards, as everyone kept telling them, "but you only compared X card, what about Y card?" This is not even an SLI vs CF test. This simply a matter of "is v-sync the answer?" type of test.

I'd also like to note, that while the average user doesn't use SLI or CF, the amount of enthusiasts who do are much higher. The ones who actually read this stuff.

 

[Imouto]

Attention whores.

It's their business and I understand that they have to generate traffic but this is beyond dumb already.

When TR started using FRAPS they didn't make such a fuss out of it. Hell, none of the other sites using FCAT is this obnoxious.

 

[bystander36]

Attention whores.

It's their business and I understand that they have to generate traffic but this is beyond dumb already.

When TR started using FRAPS they didn't make such a fuss out of it. Hell, none of the other sites using FCAT is this obnoxious.

How is it obnoxious to explorer all the aspects of multi-GPU configurations, when it has never been done before?

It sounds like you use crossfire and don't want to be reminded of the issues you have. There is a good way to do that. Don't read their articles.

 

[Riceninja]

i dont think the reviewer was purposely trying to make AMD look bad. it seems that he went with a titan SLI simply because it is as powerful as it gets and should run 2560 at 60fps seamlessly, removing any gpu constraints. no one with half a brain, or is in the market for buying gpus would compare dual titans to a single card solution from amd.

this is coming from a 7950 owner.

 

[BrightCandle]

Anybody else surprised that there are people unable to tell the difference between 30 and 60 fps? I mean wow side by side they are really different to me.

 

[zaydq]

Anybody else surprised that there are people unable to tell the difference between 30 and 60 fps? I mean wow side by side they are really different to me.

I don't. Probably because i'm more into whatever game i'm playing, and focusing on the depiction i'm seeing.

 

[VulgarDisplay]

Anybody else surprised that there are people unable to tell the difference between 30 and 60 fps? I mean wow side by side they are really different to me.

I'm still waiting for reviewers to get on the 120hz bandwagon. It's good that they are all on the smoothness bandwagon now, but they need to get with the program.

120hz is where it's at.

 

[parvadomus]

Yes 120hz would be a great test, vsync at that high speed should be almost latency-free. Maybe its the best of both worlds, zero-stutter + almost zero-latency

 

[lilrayray69]

I honestly have never been able to tell a big difference between 30-60 FPS. People say 60 is much "smoother", but quite often I can find 25 FPS in a shooter to be playable without even noticing it. I've never seen a 120hz monitor in action myself, only some comparison videos which probably don't do much justice - but still in those I couldn't tell a difference.

Same way I've never been able to see much difference with AA. Supposedly I have vision better than most, at 20/15, but maybe my eyes suck at detecting that stuff.

Also, I have to enable VSync in almost every game I play - especially shooters - because it always seems like they have horrible screen tearing and I can't stand that. So I can see screen tearing at least! But of course people in this thread say they hardly notice it and never use VSync....

I also don't notice any input lag with Vsync, lol.

 

[BrightCandle]

I notice the tearing and it impacts the image quality substantially I would rather not suffer it. Equally I notice the jaggies without AA but the performance impact is normally pretty dramatic.

I was fine up until about 2003, I was happy with 30 fps or so on a 85Hz CRT with vsync. Then I started playing more and more competitively, first Counter Strike and then a European Championship in Call of Duty. I found dropping the quality, increasing the frame rate to 60 (LCD by this point) improved my game a little and then removing vsync also further improved it. I couldn't really see the difference back then, I could feel it but not directly see it.

Fast forward nearly 10 years and the difference between 30 and 60 fps is now enormous, and I am finding 120Hz is a suitable advancement over 60hz. While I no longer play at a competitive level and haven't won a championship in years I still need that level of performance to feel the movement in the game is attached to me. I can see the animation moving at 30hz, even if its only momentary like with vsync and anything below 60fps. When vsync is on it feels detached from what I am doing, the latency from input to the movement is easily noticeable in comparison to vsync off. Thankfully I don't yet feel the need for 120 fps, a minimum of 60hz is what I am aiming for, but consistent 120hz improves my game, its just not really realistic to get at it consistently.

Ultimately I want a card that can do 120Hz reliably with decent graphics and not get the additional latency of SLI/crossfire and I also want to reduce the context queue. I don't yet know where my threshold lays but I doubt I am at it yet, games certainly don't yet feel smooth.

 

[Subyman]

Eh, I couldn't tell a difference without really looking. IMO you have to play the game and be in control to get a feeling of the difference especially since the feed has been compressed.

 

[railven]

Eh, I couldn't tell a difference without really looking. IMO you have to play the game and be in control to get a feeling of the difference especially since the feed has been compressed.

Gonna have to go with this, to a certain degree. I play one game that is so CPU bottlenecked it's scary.

Anyways, after playing this game for long, I was amazed to realize 30-40FPS felt SMOOTH. Haha. I guess my brain just accepted. It felt damn smooth.

 

[bystander36]

I notice the tearing and it impacts the image quality substantially I would rather not suffer it. Equally I notice the jaggies without AA but the performance impact is normally pretty dramatic.

I was fine up until about 2003, I was happy with 30 fps or so on a 85Hz CRT with vsync. Then I started playing more and more competitively, first Counter Strike and then a European Championship in Call of Duty. I found dropping the quality, increasing the frame rate to 60 (LCD by this point) improved my game a little and then removing vsync also further improved it. I couldn't really see the difference back then, I could feel it but not directly see it.

Fast forward nearly 10 years and the difference between 30 and 60 fps is now enormous, and I am finding 120Hz is a suitable advancement over 60hz. While I no longer play at a competitive level and haven't won a championship in years I still need that level of performance to feel the movement in the game is attached to me. I can see the animation moving at 30hz, even if its only momentary like with vsync and anything below 60fps. When vsync is on it feels detached from what I am doing, the latency from input to the movement is easily noticeable in comparison to vsync off. Thankfully I don't yet feel the need for 120 fps, a minimum of 60hz is what I am aiming for, but consistent 120hz improves my game, its just not really realistic to get at it consistently.

Ultimately I want a card that can do 120Hz reliably with decent graphics and not get the additional latency of SLI/crossfire and I also want to reduce the context queue. I don't yet know where my threshold lays but I doubt I am at it yet, games certainly don't yet feel smooth.

I never played competitively, but everything you said about different FPS is how I feel, with one added twist. When I feel heavy latency, I get nauseated quickly, and can last longer without getting nauseated up to about 80 FPS.

 

[blastingcap]

So has anyone looked at the downloaded version? Is it materially different than the Youtube version?