Jump to content

Archived

This topic is now archived and is closed to further replies.

Bobu

My ATI Results and Observations

Recommended Posts

There are a lot of posts here (and on other FSX sites) claiming that nVidia is a far better graphic card for FSX than similar ATI/AMD offerings. Seeing as I play other sims/games as well, it isn't exactly practical to change to nVidia just for FSX (and doing so could hurt other sims/games I play). So I basically assumed that I had a second-class experience with FSX, due to my graphics card. This perception changed once I saw the benchmarks people were posting here using top-of-the-line nVidia cards.

 

First, here are my system specs:

 

Windows 7 x64

Intel i7 875K (Overclocked to 4.0GHz)

8GB RAM

ASUS EAH 5850 Top DirectCU 1GB (Slight OC)

ASUS Maximus III Formula MB

SATA HD (No SSD's)

ASUS 24" 1920 x 1200 x32 LCD

 

When doing my tests, I believe my system was handicapped relative to the higher end systems for the following reasons:

 

1. Older graphics card

2. CPU not running as high as other systems (4.5 to 4.8GHz seemed to be more common)

3. Less video RAM (1GB vs 2, 3 or even 4GB)

4. Running resolution of 1920x1200 vs. 1920x1080

5. Using Orbx scenery for all tests

6. Using REX

7. Running Ultimate Traffic 2 for all tests (30% settings)

 

So given the scenario I tested under, I was expecting much worse results compared to others. Surprisingly, this was not the case.

 

Here are two videos showing my system:

 

Note: The best way to watch these videos is to click the YouTube logo in the bottom right, and then select 720p. That is as close to how I see it in-game.

 

Video 1: F1 Mustang using Orbx PNW Scenery

 

In this test I was able to get fps in the mid 20's to mid 30's without any noticeable jaggies. The video does show jaggies, except when seen in 720p (and even then I see some dancing lines I do not see on screen).

 

http://youtu.be/9Me36q-RXRI

 

Video 2: GPUMark Test

 

The second test was done using the GPUMark cloud test. Again, I had excellent performance without any noticeable jaggies. I think the 720p video also shows some jaggies, which I do not see in game.

 

http://youtu.be/Y3hsnP4eOBM

 

Based on my results, and comparing them to the higher end systems, I can only come up with a few reasons for the results I am seeing:

 

1. What I see is considerably worse than the nVidia users are seeing

 

This is possible, which is why I made the videos. It would be helpful to have nVidia users post a similar video so we can compare the results. I am thinking this could be the case, but I haven't seen any videos online that show anything demonstrably better than what I am seeing.

 

2. It takes very specific settings to unlock an ATI/AMD card's performance

 

I get the feeling that people here have thrown in the towel on ATI/AMD cards for FSX, so I am not sure how much tweaking has been done with them. In my case, I did a fair bit, and will gladly share my settings with other ATI/AMD users.

 

3. I am not comparing apples to apples

 

This is possible, and if so I would expect number 1 above to show serious differences. Maybe ATI/AMD cards do not require the same level of AA as nVidia does to reduce jaggies, and so we get the same performance without these settings? Not sure.

 

The last point needs further clarification. I have read about SGSS and the other settings nVidia users have. Doing some research, it appears that these settings help remove jaggies. Since I don't have jaggies now, I am thinking that the ATI/AMD cards deal with it another way. So unless there is some other benefit to these settings, I am not sure they really apply. nVidia users need certain settings to make things look nice, and ATI/AMD users have other settings to address the same things. But in the end, getting smooth, jaggy and tear-free visuals is all that matters IMO.

 

I am not saying that ATI/AMD is better than nVidia. I am asking for clarification on why my older ATI/AMD system is performing as well (or even better) than high-end nVidia based-systems in FSX. Maybe the visuals nVidia users see are SO MUCH better, and that is the end of it. If so, I would like to see video comparisons proving that. Claims they are better really won't do it. Or, maybe there are specific settings I use that can help other ATI/AMD users enjoy FSX more? If so, I will help as much as possible so they can get the same level of enjoyment I get.

 

One more thing. I own and use both nVidia and ATI/AMD cards. I really don't care which is better. I am not promoting either. They are just tools to make my gaming and computer experience better.

 

I look forward to your comments.

 

Rob

 

BTW, here are the images showing the fps counter during the tests:

 

7bf89662.jpg

 

0f9524c4.jpg


Rob Carr

Share this post


Link to post
Share on other sites

amd cards are not as powerful as nvidia card hence the claim of them being worse is true, well the word worse should be slower but that is why amd cards are cheaper than nvidia cards.

 

An amd 6870 runs with a few less fps than my GTX470 through tests i have done but then it is a slower and cheaper card so slower performance would be expected and there is driver issues with fsx with the amd card and the way it handles 2d images ( it might of been fixed now but i doubt it ).

 

The truth of it is Nvidia is better as its the faster card even if it costs more as it could mean an extra 5 fps :)


-Paul-

Share this post


Link to post
Share on other sites

The truth of it is Nvidia is better as its the faster card even if it costs more as it could mean an extra 5 fps :)

 

Those are nice statements, and I see them all over this forum. But without concrete proof, they are just statements IMO. I provided 2 videos using an ATI/AMD card that is 2-3 years old. The images show my fps. Can you post a video showing your results in a similar scenario?


Rob Carr

Share this post


Link to post
Share on other sites

sadly i dont have the 6870 any more so i cant do tests with it but if you are willing to take a leap of faith my 470 gave me up to 5 extra fps. It could of been the way i configured the AA and AF but im not sure and i would need the card to test as the results i get on my pc are only valid to my pc as fsx is that type of sim where you can have two almost identical pc's and the performance on them both could be worlds apart.

 

All that should matter is that the sim is smooth and stutter free and the person using what ever card they choose to buy is happy :)


-Paul-

Share this post


Link to post
Share on other sites

It could of been the way i configured the AA and AF but im not sure...

 

I am thinking that could be the case, especially since I had to tweak mine quite a bit to get the performance I wanted. And the test doesn't really need to include an ATI/AMD card. I looked up your card and it appears that yours is significantly better than mine. Could you possible do the same video tests I did with your GTX470 and upload the video so we can compare? Yours should be significantly better than mine.


Rob Carr

Share this post


Link to post
Share on other sites

I have a faulty cpu so any results i get will be subjective and may not reflect a fully working i7-920.. sry :(


-Paul-

Share this post


Link to post
Share on other sites

ATI/AMD system is performing as well (or even better) than high-end nVidia based-systems in FSX.

 

If we were to test side by side with the SAME settings in FSX and the SAME AA levels I am extremely confident sure anything above a GTX560 would blow your 5850 out of the water. [in FSX]

 

I had a 5870 a few months ago and I can definitively say that it was abysmally awful in rendering clouds in FSX compared to my slower GTX450.

This argument is not even up for debate anymore. - It is well established that AMD cards perform much slower in FSX than Nvidia. AMD dropped support for SM2 hence the horrible performance.

 

Also, you give us no indication of what your AA settings were or what your FSX settings were.

 

 

In your second photo it also looks like there is no autogen... :rolleyes:

Share this post


Link to post
Share on other sites

Hey Ben,

 

I appreciate you weighing in on this, but I have to take issue with some of your comments. First, what level of AA I am running makes zero difference. You see the video. I have almost zero jaggies. So what level I need to run to reduce it may be very different than what you run. Sure I could set my AA super high and it would make my system look bad. But I DON'T NEED TO DO THAT to get great results. Maybe you need to do that with your card to eliminate the jaggies. I don't. Did you not like the clouds in the video? I thought they looked sweet, and you see the performance. So maybe the settings with your 5870 were incorrect? No idea, but the proof is in the videos.

 

As far as autogen goes, I had it set to the Orbx recommended settings. I believe it was Normal.

 

What I don't understand is the emphasis on SAYING nVidia is superior to ATI/AMD for FSX without providing any proof. Ben, I put the videos up in the hope that nVidia users would show me (and others) WHY nVidia is better. So far all I get are CLAIMS without any proof. Saying it is "well established that AMD cards perform much slower" does not make it true. You see the videos. My system is no where near the specs of yours. Explain to me why your system does not outperform mine? I am hoping you can.

 

Otherwise, maybe I have stumbled onto settings that eliminate the performance difference. Considering that ATI cards are cheaper, and they handle multiple displays way better (until recently, nVidia had no answer for eyefinity), this could be significant for others.

 

So instead of saying that nVidia is better, why not SHOW us? You see my specs. You see the videos. Either someone is going to clearly show the difference, or we can conclude the ATI may in fact be the better cards, for a LOT less money. And personally, I haven't made up my mind yet. I am still waiting for the proof...any proof that shows nVidia beats a well tuned and balanced ATI/AMD based system.

 

Thanks for chiming in.

 

P.S. I am really just hoping someone...anyone...can show where the difference is. I will gadly admit that nVidia is better for FSX if I can see it.


Rob Carr

Share this post


Link to post
Share on other sites

With the exact the same system in my profile, i had an ATi 7950 3GB, the XFX Black edition. Followed Word Not Allowed´s guide step by step, and, with the exact the same settings on Word Not Allowed´s guide, i was getting 12 FPS in his cloud test with Super Sampling enabled (in order to equal nvidia´s SGSS quality).

 

I sold the card for 220€, added 180 and got an eVGA 670 FTW and overclocked it.

 

Now, 27 FPS, more than 100% gain.

 

In all other games (pCARS,rFactor 2, iRacing, etc), the card was fine, but i think (well, i´m sure now after lose a lot of money), that AMD card can´t compete with nvidia on AA performance, simply as that.

 

Each system is defferent, i´m talking about my personal experience, if you DO NOT use FSX, then AMD cards are the best bang for the buck maybe, but for FS in general, it just does not work with an "minimal" AA for avoid shimmer.

Share this post


Link to post
Share on other sites

Show me your CCC settings.

 

I will post it in a few minutes.

 

With the exact the same system in my profile, i had an ATi 7950 3GB, the XFX Black edition. Followed Word Not Allowed´s guide step by step, and, with the exact the same settings on Word Not Allowed´s guide, i was getting 12 FPS in his cloud test with Super Sampling enabled (in order to equal nvidia´s SGSS quality).

 

I sold the card for 220€, added 180 and got an eVGA 670 FTW and overclocked it.

 

Now, 27 FPS, more than 100% gain.

 

In all other games (pCARS,rFactor 2, iRacing, etc), the card was fine, but i think (well, i´m sure now after lose a lot of money), that AMD card can´t compete with nvidia on AA performance, simply as that.

 

Each system is defferent, i´m talking about my personal experience, if you DO NOT use FSX, then AMD cards are the best bang for the buck maybe, but for FS in general, it just does not work with an "minimal" AA for avoid shimmer.

 

I guess I am wondering why my 5850 is outperforming your system.

 

One thing does come to mind. It looks like a lot of the tests are done with 1920x1080 resolution. I am running 1920x1200. Maybe the additional resolution negates the need for higher AA settings? If that is the case, that could be significant information for other users, especially those with ATI cards.

 

Show me your CCC settings.

 

Any chance I will ever see a video doing the same test with your system, Ben?


Rob Carr

Share this post


Link to post
Share on other sites

I'm confused as to what we're getting at in this thread. No sarcasm, I just need someone to actually say what it is we're trying to accomplish here.

 

To the OP. You specifically admit you're comparing apples to oranges. You offer no AA settings used, no scenario made specifically for the "comparison". No standard is followed. I'm just lost! :blush:


___________________________________________________________________________________

Zachary Waddell -- Caravan Driver --

Facebook: http://www.facebook.com/zwaddell

Avsim ToS

Avsim Screenshot Rules

Share this post


Link to post
Share on other sites

 

I will post it in a few minutes.

 

I guess I am wondering why my 5850 is outperforming your system.One thing does come to mind. It looks like a lot of the tests are done with 1920x1080 resolution. I am running 1920x1200. Maybe the additional resolution negates the need for higher AA settings? If that is the case, that could be significant information for other users, especially those with ATI cards.

 

Any chance I will ever see a video doing the same test with your system, Ben?

 

Sorry, with my settings in fsx(exactly the same as Word Not Allowed's) and his cloud test, with all my respects, is totally impossible that you can get 27 fps on the aa cloud test. Of course with the same fsx settings and comparable levels of aa of 8s+2 or 4 sgsaa.

 

For me, Word Not Allowed's cloud test is a very good benchmark of what a gpu is capable of, follow his guide and settings, do the cloud bench. Im almost 100% sure that it will be difficult for a 5850 to get double digits fps, based on performance of my ex 7950.

 

Tapatalk2, Samsung Galaxy Note.

Share this post


Link to post
Share on other sites

Did you not like the clouds in the video?

 

Didn't see your vids... but I had a 3870x2... as you say, clouds were ooh-la-la fantastic but fps a slideshow particularly with lots of clouds. I rather switch than fight... so I run a GTX 580.

 

You might want to also start a thread on this (if not finding answers in search) over at SimForums in the hardware/software/computer technology forum http://www.simforums...orum29.html.��A guy by the nick of "NickN"... I know for a fact Nick could answer your question(s) about this...

Share this post


Link to post
Share on other sites

Any chance I will ever see a video doing the same test with your system, Ben?

 

I am done with this... This thread has accomplished nothing.

 

It clearly shows that you come in here, try to disgruntle a few members with some bold statements, and then attempt to cover your ground without any solid basis.

 

You offer no AA settings used, no scenario made specifically for the "comparison". No standard is followed.

 

+1

Share this post


Link to post
Share on other sites

  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
  • Donation Goals

    AVSIM's 2020 Fundraising Goal

    Donate to our annual general fundraising goal. This donation keeps our doors open and providing you service 24 x 7 x 365. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. We reset this goal every new year for the following year's goal.


    22%
    $5,540.00 of $25,000.00 Donate Now
×
×
  • Create New...