Asking Ryan Smith of AT if a special examination could be done?

[Vesku]

In regards to Skyrim, here are my results with Fraps. 12.11 Beta 8 drivers:

 

[thilanliyan]

^Can you overlay the frames per second over that?

 

[SlowSpyder]

Keys, I like your posts and am happy you are part of these forums, Nvidia focus group or not. I think often you get flak when it is completely not deserved. But I can't help but feel like this is an attempt to get Ryan to write an article that will show Nvidia in a better light than the competition. Just look at it from my perspective, a poster associated with Nvidia trying to have an article written that more or less shows frame rates don't matter, Nvidia's smoothness matters at a time when Nvidia is producing lower frame rates at most every price point (we all know how the internet works, "zomg! Nvidia is smoother despite lower frame rates!" and people will think that this applies to single cards, too).

Maybe I'm 100% wrong here, it is an article I would read, but it seems odd that this request was made when there really isn't too big of a 'war' over this.

 

[lavaheadache]

I think it is worth an investigation. I want to know if there is infact a "real" difference.

 

[SolMiester]

Whats wrong with an article that shows NV in a better light than AMD if true?....Not like these are AMD forums.......is it?

 

[WaitingForNehalem]

I would be interested in this.

 

[Zanovar]

Yeah, it's all just me Willie. I'm at war with myself. I was always a self-conflicted individual longing for a sense of smoothness to my gaming experiences. It isn't a focus group thing. It's my thing. :thumbsup:

I highly doubt that,as lava mentions its worth investigating.

 

[jimhsu]

Some results for skyrim:

This is a fairly maxed out ENB setup with highres textures+the 2K pack.

2500K
16GB RAM
2GB 6950 (905/1375 unlocked)
Windows 8 / 12.11 beta 11




Exteriors are barely playable, but I mainly use Skyrim for screenshots/machinima.

 

[lavaheadache]

Skyrim needs a frame cap. I had that problem with my 580 too. Simply forcing a frame limit at 60 fps fixes that stuttering. I repeat, I had the same problem with my 580 and I have it now with my 7970.

 

[Ryan Smith]

This is something we will eventually look at, but as I've mentioned before it's not something I want to do until I have the right tools (which I am expecting sooner than later). Scott has done an amazing job with FRAPS, but that NVIDIA quote isn't wrong.

To really go at this I'd like to be able to time frame updates at a low level (i.e. the actual buffer swaps), but also keep track of a frame's time in the rendering pipeline. Ideally we need timestamps both for the buffer swap and the simulation itself. This is to check the rate at which frames are being displayed and the rate they're being generated. Both of these factors need to be consistent if you want to maximize smoothness. E.G. if frames are being displayed every 30ms but the simulation is generating them at a far more variable rate (say 15ms and 45ms repeating) then it's not going to be particularly smooth, just as if the buffer swaps are taking place at an uneven interval.

In the meantime however I'm personally satisfied with both AMD and NVIDIA in this regard; in my own personal experience I don't believe either of them to be notably worse than the other when it comes to single-GPU configurations. At the same time this community has a terrible habit of making mountains out of molehills, so if there's a "big war" brewing then I fear you guys might be taking this whole subject a bit too seriously. Spend less time looking at charts and more time playing video games, it's not like there's a shortage of good action games this year.

 

[chimaxi83]

This is something we will eventually look at, but as I've mentioned before it's not something I want to do until I have the right tools (which I am expecting sooner than later). Scott has done an amazing job with FRAPS, but that NVIDIA quote isn't wrong.

To really go at this I'd like to be able to time frame updates at a low level (i.e. the actual buffer swaps), but also keep track of a frame's time in the rendering pipeline. Ideally we need timestamps both for the buffer swap and the simulation itself. This is to check the rate at which frames are being displayed and the rate they're being generated. Both of these factors need to be consistent if you want to maximize smoothness. E.G. if frames are being displayed every 30ms but the simulation is generating them at a far more variable rate (say 15ms and 45ms repeating) then it's not going to be particularly smooth, just as if the buffer swaps are taking place at an uneven interval.

In the meantime however I'm personally satisfied with both AMD and NVIDIA in this regard; in my own personal experience I don't believe either of them to be notably worse than the other when it comes to single-GPU configurations. At the same time this community has a terrible habit of making mountains out of molehills, so if there's a "big war" brewing then I fear you guys might be taking this whole subject a bit too seriously. Spend less time looking at charts and more time playing video games, it's not like there's a shortage of good action games this year.

Exactly, very well said Ryan. I use both AMD and Nvidia, single and multi GPU, and haven't noticed any appreciable difference... well, other than my 7970 being faster, of course. The problem here is that these guys are taking what's said as gospel, without having actually used the hardware themselves. Using the hardware yourself is important because everybody sees things differently. Some see tearing, some don't. Some see stutter, some don't. This isn't black and white.

Anyway, I'd be interested in an actual technical investigation into this, not just "it just feels like it", because that means nothing (unless you're a diehard Nvidia fan).

 

[Vesku]

A new run with FPS data, although since Skyrim physics appears tied to Vsync I find it amusing Techreport uses it for it's smoothness testing:

I'd be interested more in this "smoothness" testing if it was done with high fps cameras and multiple monitors repeated for lots of different GPUs. I personally have noticed no stuttering with my current GPU and I seem moderately sensitive to the phenomenon based on my dual GPU experiences. As of now I'm going to side with the mountains out of molehills viewpoint.

 

[Hitman928]

In regards to Skyrim, here are my results with Fraps. 12.11 Beta 8 drivers:

Is this outside or inside, any fighting?

 

[Vesku]

Outside, I did soultrap a few mudcrabs along my path. Both runs, second run also had a mountain lion.

 

[blastingcap]

This is something we will eventually look at, but as I've mentioned before it's not something I want to do until I have the right tools (which I am expecting sooner than later). Scott has done an amazing job with FRAPS, but that NVIDIA quote isn't wrong.

To really go at this I'd like to be able to time frame updates at a low level (i.e. the actual buffer swaps), but also keep track of a frame's time in the rendering pipeline. Ideally we need timestamps both for the buffer swap and the simulation itself. This is to check the rate at which frames are being displayed and the rate they're being generated. Both of these factors need to be consistent if you want to maximize smoothness. E.G. if frames are being displayed every 30ms but the simulation is generating them at a far more variable rate (say 15ms and 45ms repeating) then it's not going to be particularly smooth, just as if the buffer swaps are taking place at an uneven interval.

In the meantime however I'm personally satisfied with both AMD and NVIDIA in this regard; in my own personal experience I don't believe either of them to be notably worse than the other when it comes to single-GPU configurations. At the same time this community has a terrible habit of making mountains out of molehills, so if there's a "big war" brewing then I fear you guys might be taking this whole subject a bit too seriously. Spend less time looking at charts and more time playing video games, it's not like there's a shortage of good action games this year.

...or, just use a high speed camra.

 

[Zanovar]

...or, just use a high speed camra.

hahah

 

[badb0y]

Ahh is this including single cards or dual card configs only?

I thought the whole fuss was that nVidia has some sort of hardware implementation that makes SLi smoother I didn't think it included single card configs too. I'm pretty satisfied with my card but I would love to know if the gameplay is smoother on nVidia cards.

 

[Hitman928]

Outside, I did soultrap a few mudcrabs along my path. Both runs, second run also had a mountain lion.

I think this helps confirm my suspicions that tech reports new numbers seem to be more an issue with their new test bench (windows 8 computer) as your results match what they were getting before and you use a more demanding run.

 

[Vesku]

I think this helps confirm my suspicions that tech reports new numbers seem to be more an issue with their new test bench (windows 8 computer) as your results match what they were getting before and you use a more demanding run.

Could be a Windows 8 thing, I'm sticking with Windows 7. Honestly, Techreport and HardOCP need to do more thorough investigations. Disservice to the community to bring up "smoothness" of gameplay and not delve deeper.

 

[bononos]

Keys, I like your posts and am happy you are part of these forums, Nvidia focus group or not. I think often you get flak when it is completely not deserved. But I can't help but feel like this is an attempt to get Ryan to write an article that will show Nvidia in a better light than the competition. Just look at it from my perspective, a poster associated with Nvidia trying to have an article written that more or less shows frame rates don't matter, Nvidia's smoothness matters at a time when Nvidia is producing lower frame rates at most every price point (we all know how the internet works, "zomg! Nvidia is smoother despite lower frame rates!" and people will think that this applies to single cards, too).

Maybe I'm 100% wrong here, it is an article I would read, but it seems odd that this request was made when there really isn't too big of a 'war' over this.

It still makes sense to clear up the issue of whether Radeon cards have more jitter whether or not it is due to fanboy partisan motives.

There was one post in some thread here at AT which I remember where a veteran (or semi vet) poster claimed smoother gameplay with Nvidia vs Radeon non-sli. Edit- that was before I saw some other newer threads in this forum with active discussions on the jitter issue.

 

[jimhsu]

Windows 8 + old drivers probably. So far 12.11b11 is working decently on Windows 8.

Yes I'm frame capped, but in normal play my game runs far too slow to even hit the cap. Looking to upgrade to 7xx or 89xx, but I'll wait.

 

[Rvenger]

This is something we will eventually look at, but as I've mentioned before it's not something I want to do until I have the right tools (which I am expecting sooner than later). Scott has done an amazing job with FRAPS, but that NVIDIA quote isn't wrong.

To really go at this I'd like to be able to time frame updates at a low level (i.e. the actual buffer swaps), but also keep track of a frame's time in the rendering pipeline. Ideally we need timestamps both for the buffer swap and the simulation itself. This is to check the rate at which frames are being displayed and the rate they're being generated. Both of these factors need to be consistent if you want to maximize smoothness. E.G. if frames are being displayed every 30ms but the simulation is generating them at a far more variable rate (say 15ms and 45ms repeating) then it's not going to be particularly smooth, just as if the buffer swaps are taking place at an uneven interval.

In the meantime however I'm personally satisfied with both AMD and NVIDIA in this regard; in my own personal experience I don't believe either of them to be notably worse than the other when it comes to single-GPU configurations. At the same time this community has a terrible habit of making mountains out of molehills, so if there's a "big war" brewing then I fear you guys might be taking this whole subject a bit too seriously. Spend less time looking at charts and more time playing video games, it's not like there's a shortage of good action games this year.

My thoughts exactly. My experience has also been good with both cards. In the end it just comes down to price for me.

 

[Eureka]

This is something we will eventually look at, but as I've mentioned before it's not something I want to do until I have the right tools (which I am expecting sooner than later). Scott has done an amazing job with FRAPS, but that NVIDIA quote isn't wrong.

To really go at this I'd like to be able to time frame updates at a low level (i.e. the actual buffer swaps), but also keep track of a frame's time in the rendering pipeline. Ideally we need timestamps both for the buffer swap and the simulation itself. This is to check the rate at which frames are being displayed and the rate they're being generated. Both of these factors need to be consistent if you want to maximize smoothness. E.G. if frames are being displayed every 30ms but the simulation is generating them at a far more variable rate (say 15ms and 45ms repeating) then it's not going to be particularly smooth, just as if the buffer swaps are taking place at an uneven interval.

In the meantime however I'm personally satisfied with both AMD and NVIDIA in this regard; in my own personal experience I don't believe either of them to be notably worse than the other when it comes to single-GPU configurations. At the same time this community has a terrible habit of making mountains out of molehills, so if there's a "big war" brewing then I fear you guys might be taking this whole subject a bit too seriously. Spend less time looking at charts and more time playing video games, it's not like there's a shortage of good action games this year.

It's the Lord of the Flies mentality. Put two camps in a closed community and they will go to war over the smallest thing. People will go to war for anything.

 

[Termie]

I always think it's amazing that Ryan stops by the forums to respond to all of the requests from Anandtech readers. It's even more amazing that he only has 135 posts in 7 years...


Maybe we're not being hard enough on him.

 

[atticus14]

only interested if Radeon Pro tools are involved