PS: waiting for more screenies on other systems!

 

typo i5 of course.

 

In manhattan scenery I get around 12-15 frames everything max. I will post a snapshot later today.

i'm no where near those 17 FPS in real practical gameplay, that must be a very picky flight scenario to try to prove nvidia runs it better...

 

If you are never in this picky scenario, and as I mentioned previously, check that realtime weather is off, because something ain't right, then you should consider yourself lucky. I'm very often in such scenario. And I think many others are. It's enough that I fly NGX into LOWI and have a bad weather, FPS are sinking into low 20s in VC.

ELNIKA,

 

Since you are so happy with your card, I would really like to know how it performs in cloud situations.

Would you please be so kind and load this flight: http://dl.dropbox.co...6/cloudtest.rar

Press S once to enter external view, and by using FRAPS (google it, its free for FPS measurement), note the FPS.

It's vital that the measurement is via FRAPS and that its taken in the external view, exactly how I placed it. Just press S once after flight has loaded. Its default cessna so you should have it.

 

Then, if you came this far: make a screenshot of the aircraft closeup and post it here for quality comparison.

 

My results were as follows :

Catalyst SuperSampling AA mode : 4x (edge detect)

Outside view

Right wing view:

VC View:

Hope that gives you an idea.

To answer your previous question, these were exactly the kind of tests I made when I was comparing my 570GTX vs my HD6970. IQ was the same (for as far as I could tell...) and believe me or not out of the several ten's of tests I made I can count on only one hand the times that I saw a higher framerate on the 570GTX.

Besides that I must say that the game just felt (<-- yes I know this is very subjective as you cannot proof a feeling ^.^) smoother with my ATI, it came to little things really (some microstutters that did not happen with the ATI and consistently came up with the 570GTX).

I must honestly say that I am still wondering if maybe my tweaking for the GTX was not optimal but I'll never really know I guess... anyway I am having the time of my life (FSX-wise ;p) with the HD6970 and when the time comes that I need an upgrade it could be an Nvidia if it turns out that (overal) it is the better card for me...

 

cheers

I must say, the image quality appears to be quite good, no jaggies on the aircraft lines, though quite hard to judge, since the pic is smaller.

 

But, what baffles me, are the clouds. You loaded my situation apparently, but you have no overcast. Did you have any cloud engine OR FSX realtime clouds enabled while doing that? I am sure FSX loaded other weather situation for this file than my original, thus giving you completely false result. You wouldn't possibly see blue sky in my situation.

 

Actually i have REX weather engine ON with a whole custom Theme on FSX, i dont know if that will make a difference :S

My results were as follows :

Catalyst SuperSampling AA mode : 4x (edge detect)

 

Outside view

 

Right wing view:

 

VC View:

 

Hope that gives you an idea.

To answer your previous question, these were exactly the kind of tests I made when I was comparing my 570GTX vs my HD6970. IQ was the same (for as far as I could tell...) and believe me or not out of the several ten's of tests I made I can count on only one hand the times that I saw a higher framerate on the 570GTX.

Besides that I must say that the game just felt (<-- yes I know this is very subjective as you cannot proof a feeling ^.^) smoother with my ATI, it came to little things really (some microstutters that did not happen with the ATI and consistently came up with the 570GTX).

I must honestly say that I am still wondering if maybe my tweaking for the GTX was not optimal but I'll never really know I guess... anyway I am having the time of my life (FSX-wise ;p) with the HD6970 and when the time comes that I need an upgrade it could be an Nvidia if it turns out that (overal) it is the better card for me...

 

cheers

 

Nice results buddy!!, now remember, you have a much much more powerfull processor than mine, that should help you deal with lot of situations!!, Ivy Bridge cant come soon enough :P.

 

I will soon post more screens on my driver settings and FSX settings for comparison purposes.

As promised, here are my settings screens:

 

 

 

 

 

I also disabled REX weather engine and Vatsim weather, did a test again, but no difference on the clouds:

 

 

 

 

Cheers

nice clouds , are they REX2 ?

My results were as follows :

Catalyst SuperSampling AA mode : 4x (edge detect)

Outside view

Right wing view:

VC View:

Hope that gives you an idea.

To answer your previous question, these were exactly the kind of tests I made when I was comparing my 570GTX vs my HD6970. IQ was the same (for as far as I could tell...) and believe me or not out of the several ten's of tests I made I can count on only one hand the times that I saw a higher framerate on the 570GTX.

Besides that I must say that the game just felt (<-- yes I know this is very subjective as you cannot proof a feeling ^.^) smoother with my ATI, it came to little things really (some microstutters that did not happen with the ATI and consistently came up with the 570GTX).

I must honestly say that I am still wondering if maybe my tweaking for the GTX was not optimal but I'll never really know I guess... anyway I am having the time of my life (FSX-wise ;p) with the HD6970 and when the time comes that I need an upgrade it could be an Nvidia if it turns out that (overal) it is the better card for me...

 

cheers

 

What were the AA settings you tried in the GTX570 that looked comparably as good as that in the 6970?

ELNIKA - Try setting your 'Cloud Detail' to MAX and you should get full coverage... I get 22fps with this save running 4096 REX clouds and near max settings... need standard settings really to compare. Its a tough test alright, with five layers of Cumulus, and not realistic, as real wolrd Cumulus dosn't work like this [real world as in reality, not FSX weather set-up]

 

I'm certainly not going to say AMD will outpace nV at the same settings, but I certainly do not get a slide show, even at these mad cloud levels.

 

It would be interesting to see some nV results... might just be tempted to get a 680 [AMD drivers P me off] :Party:

nice clouds , are they REX2 ?

 

Yes, they are REX 2 xD

 

ELNIKA - Try setting your 'Cloud Detail' to MAX and you should get full coverage... I get 22fps with this save running 4096 REX clouds and near max settings... need standard settings really to compare. Its a tough test alright, with five layers of Cumulus, and not realistic, as real wolrd Cumulus dosn't work like this [real world as in reality, not FSX weather set-up]

 

I'm certainly not going to say AMD will outpace nV at the same settings, but I certainly do not get a slide show, even at these mad cloud levels.

 

It would be interesting to see some nV results... might just be tempted to get a 680 [AMD drivers P me off] :Party:

 

Sure, will try later ;)

What were the AA settings you tried in the GTX570 that looked comparably as good as that in the 6970?

 

Now you're asking for something which I didn't look back at for more than 6 months now. It was either the 8xS with 4x supersampling or the 16xS with 2x supersampling, but for the life of me I cannot be 100% certain anymore...

 

ELNIKA - Try setting your 'Cloud Detail' to MAX and you should get full coverage... I get 22fps with this save running 4096 REX clouds and near max settings... need standard settings really to compare. Its a tough test alright, with five layers of Cumulus, and not realistic, as real wolrd Cumulus dosn't work like this [real world as in reality, not FSX weather set-up]

 

I'm certainly not going to say AMD will outpace nV at the same settings, but I certainly do not get a slide show, even at these mad cloud levels.

 

It would be interesting to see some nV results... might just be tempted to get a 680 [AMD drivers P me off] :Party:

 

I fully agree with the fact that this should rather be considered as a though test in order to have an idea of what your card can handle. Honestly setting overcast with only 3 layers of clouds instead of 5 won't ruin the level of immersion and realism but it will definitely save up on the resources.

 

Just out of curiosity I would like to see what a GTX580 with let's say a 4.5-4.7'ish CPU would crank up on this particular situation knowing that the price tag for that card is sensibly higher ;p

My results were as follows :

Catalyst SuperSampling AA mode : 4x (edge detect)

 

You my friend have jaggies, which are obvious on thin lines that connect the fuselage and the wing.

And judging by that, I can only guess your NGX (if you own it) cockpit looks very shimmery.

 

But I don't this to become an ATI war. If you are happy with your card, we are too happy!

 

Actually i have REX weather engine ON with a whole custom Theme on FSX, i dont know if that will make a difference :S

 

Of course that makes a difference. But I see you changed that already.

 

But as someone pointed out: set full coverage in FSX. That is probably a culprit.

 

The problem with this test situation, it was never ment to test the CLARITY of the picture. It was never created to compare AA settings. It was only made to compare performance at various AA settings.

Thus, I have no way to compare how good your AA is, which you are using on ATI.

 

But let me see if I can find something useful...

You my friend have jaggies, which are obvious on thin lines that connect the fuselage and the wing.

And judging by that, I can only guess your NGX (if you own it) cockpit looks very shimmery.

 

But I don't this to become an ATI war. If you are happy with your card, we are too happy!

 

 

Sorry for the late response to this as I've been a bit busy lately.

Just to make sure that our standards about 'shimmering' textures is somewhat on the same level. In this movie we see what I would call shimmering textures in VC mode, I guess due to AA being too low or because no supersampling is being used maybe, I don't know for sure (watch it preferably at highest resolution and in fullscreen mode in order to see the shimmering problem). Anyway I wouldn't be able to play in such a VC for more than 10 minutes.

My NGX's VC textures are not shimmering (to my standards...)

 

I will try to make a short movie of my VC (when I find some time though, maybe somewhere this coming weekend)

 

http://www.youtube.com/watch?v=fB9cG3xXJpY&feature=endscreen&NR=1

 

No problem about ATI vs Nvidia war mate (I had Nvidia in the past and probably will have it in the future too, who knows) that's not my intention at all, it's just that everytime that this kind of discussion starts, it usually ends with something sounding a bit like your last sentence. Maybe it's just me but it sounds a bit ironic with the understatement saying "If you're happy with low quality, it's good for you..." and honestly I don't think my actual performance has anything to envy to what Nvidia cards have to offer (at least not what the 570GTX had to offer ;p)

 

cheers

"If you're happy with low quality, it's good for you..."

 

Don't turn my words -_- It was not at all what I ment, I ment exactly what I said! :Peace:

Just to muddy the waters a little bit without starting a new topic (because I know people still wonder which is better), I recently did some testing.

 

I do think the new 6000 series of ATI cards killed off the idea that Nvidia is better. I tested a 2GB 6950 and a GTX570. By better I mean definitively better.

 

FPS, I'm seeing almost no difference in normal situations (busy airports, broken skies, clear skies, at KSEA with Orbx, etc.). Both seem to hold within 1-2 frames of each other. Now Nvidia does seem to handle clouds better in full overcast situations. That would be one plus for Nvidia. I saw more framerate drops with the ATI card in these situations.

 

As far as image quality, a commonly cited plus for Nvidia the past few years, I didn't see any difference after extensive testing matching up. Using 8x2 + 2XSGSS in the Nvidia card produced practically identical image quality to ATI using 8xS with supersampling (I actually found ATI to look slightly better since it was doing 8x full supersampling). I was able to get AF to match up perfectly as well using 16x on both cards and crispness seemed identical on further off textures.

 

I think the ATI cards much higher memory total allows it to run higher pure AA settings yet maintain the performance levels of the 570's combined AA settings.

 

And just to clarify, at these matched up settings for image quality, performance was again practically identical as cited at the beginning of this post. So it's not like I'm sacrificing image quality to get these results on one card.

 

I was able to successfully get vsync working well on both (had no use for the NVidia 1/2 refresh tweak because I don't normally maintain 30FPS in the NGX at busy airports which is a lot of my flying. I don't think anyone does).

 

Where I found the ATI card better was artifacts with the BP=0 tweak (vital for me). It seemed, as cited by another poster in this thread to handle it much better. With the 570 I get flashing textures even in low load situations and black squares appearing sometimes. Not always, but enough to be an annoyance. This is with FFF at .10. Same settings on both cards. I also found the ATI card to be smoother overall with less microstutters.

 

Lastly, I'm getting crashes (fully lockups) with the 570 in heavy cloud situations sometimes, despite it handling the FPS better in these situations.

 

In short, ATI cards have come a long way since 2-3 years ago when it was absolutely true to say Nvidia was better for FSX. Most of the tests that prove Nvidia is better are outdated and irrelevant today though, so you have to take that into account.

 

Both cards have pros and cons. Get whatever you can get the best deal on IMO because either will work the same in 99% of situations. From there it may just be about how your system plays with each card. If one gives you crashes and the other doesn't, go with the other.

 

Also, if you are big on overcast flights, go with the Nvidia. You will see better performance from it.

 

P.S. This is with the Shader 3.0 mod installed, which does make a big difference for ATI cards. So if you have an ATI card, that's a must to achieve what I'm asserting.

This was/is a super useful discussion, many thanks to the OP and all contributors. I have a HD 5870 as you can see below, but my main performance throttle is the dual core Phenom @ 3.8 Mhz, and of course an i5/7 would help me a great deal. However, my 5870 seems to crank out 20/30 fps all settings maxed in the kind of flying I do, low level VFR and gliding/soaring close to and in cloud and over photo-scenery with auto-gen trees, but Oh Dear, a very annoying micro-stutter apparent in turns that I just can't seem to kill.

 

I use SMAA in DX9, (no FSX or CCC anti-aliasing at all) which gives me a lovely crisp display, and a Bojote-tuned fsx.cfg, and manual-install Shader mod 3 with the original terrain folder (Bojote's (bless him) waterconstants.xml is truly porked for me).

 

I do wish ATI would implement a driver-based frame rate limiter.

 

I was seriously thinking of going GTX 580 with the driver frame rate limiter in the hope to cure the stutter, but now I'm not too sure. Maybe I'll go another year with this rig and then upgrade the mobo/cpu and the vid card at the same time. Again, super informative posts, thanks to all concerned.