R600vsG80 Benchmark Comparisons From Any Game Welcome!

[Obie]

Nope, AA didnt work for me too. You can overwrite driver with these, no need to uninstall.
Hot fixes install exactly the same as Cat release. I suggest that you uninstall the old and install the Hot Fix as a new Cat release. In reality a hot fix is only suppose to remove and replace driver files, registry commands, etc. I did not see that out of the hot fix that was no different then a complete install.

[spajdr]

Well i always overwrite everything and there were never any problems :) maybe with enabling Crossfire under CCC, but it was CCC fault, not install/uninstall/reinstall process.

[spajdr]

If AA is enabled under DX10 version, it crashes the game.

[spajdr]

They are also new drivers for XP, i mean hotfix ones
hotfix_xp32_8-39-5-070625a-49541.exe

Two days ago it was
hotfix_xp_8-383-2-070622a-49428.exe

http://support.ati.com/ics/support/default.asp?deptID=894&task=knowledge&questionID=28188

[Xion X2]

If AA is enabled under DX10 version, it crashes the game.
They need to get this crap fixed, pronto.

[Xion X2]

They are also new drivers for XP, i mean hotfix ones
hotfix_xp32_8-39-5-070625a-49541.exe

Two days ago it was
hotfix_xp_8-383-2-070622a-49428.exe

http://support.ati.com/ics/support/default.asp?deptID=894&task=knowledge&questionID=28188
Do these have anything to do w/ Lost Planet?

[mynakedrat]

aa works fine on my 290xt, vista dx ten
i can put it on wide temporal 6 or eight.
or use in game settings, same thing.
BUT................
If i turn motion blur on high or concurrent rendering, then i get these little tiny black boxes on txtrs, shadows and snow.
so i turned motion blur to medium ,got rid of it.

[mynakedrat]

i have attached my stress test.
aa 6x temporal, 1680x1050, all settings on high, all settings eabled(hdr v-sync) high quality

2900xt pwns!!!!!!!!!!!!!!!!!

[Xion X2]

aa works fine on my 290xt, vista dx ten
i can put it on wide temporal 6 or eight.
or use in game settings, same thing.
BUT................
If i turn motion blur on high or concurrent rendering, then i get these little tiny black boxes on txtrs, shadows and snow.
so i turned motion blur to medium ,got rid of it.
Ok, thanks for the reply fellow 2900 owner. Looks like it's about time for me to move to Vista as ATI's drivers look better there.

[Obie]

Ok, thanks for the reply fellow 2900 owner. Looks like it's about time for me to move to Vista as ATI's drivers look better there.
I've been thinking the same thing myself. However, there isn't enough games out as of yet.

[Obie]

Has anyone benchmarked COJ with the patch (http://www.fileshack.com/file.x/10668/Call+of+Juarez+1.1.1.0+Patch+-+US) update?

[Blacklash]

Hey Xion, what do you do in the "Sanctuary demo" with no AA on, 16xAF @ 1680x and your 900MHz core?

[Xion X2]

Hey Xion, what do you do in the "Sanctuary demo" with no AA on, 16xAF @ 1680x and your 900MHz core?
I'll test it out in a bit.

For anyone who wants to see a very detailed FRAPS benchmark run w/ recorded framerates (min, avg, and max) w/ Test Drive on my 2900XT, you can go here to the following link. I'm averaging 58FPS with my current overclock at 1680x1050, max settings, 4xAA, 16xAF & HDR on the Cat 8.39s:

http://www.rage3d.com/board/showpost.php?p=1335010657&postcount=321

Fair warning, that thread is a little heated. Some blowhard know-it-alls over there have been calling BS on my screenshots, so I had to put them in their place. I tend to get wound up when these types want to twist the facts around.

[Razor1]

I'll test it out in a bit.

For anyone who wants to see a very detailed FRAPS benchmark run w/ recorded framerates (min, avg, and max) w/ Test Drive on my 2900XT, you can go here to the following link. I'm averaging 58FPS with my current overclock at 1680x1050, max settings, 4xAA, 16xAF & HDR on the Cat 8.39s:

http://www.rage3d.com/board/showpost.php?p=1335010657&postcount=321

Fair warning, that thread is a little heated. Some blowhard know-it-alls over there have been calling BS on my screenshots, so I had to put them in their place. I tend to get wound up when these types want to twist the facts around.


since you want to bring it over here, lets continue :)

you said 3 fps isn't a big deal, well you know what 3 fps is 5%-7% on your end, now 3 fps on an another benchmarks system is another 5%-7%, thats a max of a 10%-14% variation, thats quite large.

Fair warning there was only one person talking sh*t at rage, and that was you, so don't start with the smack talk here, its getting old.

[Xion X2]

but you were under the assumption TDU the frame rates were similiar in every area, they are not, and they vary widely.

I'll just take one of your tests are rip it apart, your TDU tests are crap, they are not in the same location, thats a big no no, you have different models in the scene.
http://www.rage3d.com/board/showpost.php?p=1335009391&postcount=264


http://i11.tinypic.com/66o9rfb.jpg

55.5, 57.5, 58.1. Three completely different locations/races w/ different car models. A variation of 2.6fps.

:rolleyes:

You've been owned. Just walk away.

[Razor1]

you have no idea about stats do you? I guess you must of failed CS, because your rudimentary math and analytical skills are poor :thumbdwn:

error isn't only on your side of the fence. Then you also have to look at frame rates with in each race, what is variation of your frame rates vs another test system frame rates.

[Xion X2]

you have no idea about stats do you? I guess you must of failed CS, because your rudimentary math and analytical skills are poor :thumbdwn:

error isn't only on your side of the fence. Then you also have to look at frame rates with in each race, what is variation of your frame rates vs another test system frame rates.


http://i11.tinypic.com/66o9rfb.jpg

[quote=razor1]but you were under the assumption TDU the frame rates were similiar in every area, they are not, and they vary widely.

:rolleyes:

You can't argue with an average taken from a testbed of 6 different races at 3 different locations.

You have a real problem accepting facts that go against your own mindset, you know that? You say I have the problem? You asked me to post a fraps benchmark run and that that would tell the story. Well there it is, now you just need to tuck your little tail between your legs, eat your serving of crow, and walk your @ss on back home.

[Razor1]

there are no facts there thats the problem, you are presuming there is, but no with out standerization, you can't have theory as fact, it will always be theory. Its not even theory its guess work. If you follow the lines of logic you end up with too much variation in the test cases to make a good judgement. Did the others do the same tests you did, with the same number for frames, same time runnings? There is no way I can make that assumption.

variation is not only from your system, if you understood that then you would have your friends do the exact same tests as you. Thats why we called it fud over at rage, and its still fud without the other data, I already said it should end up close in this game anyways, variation from what becco and I did was there because we weren't able to do the race exactly the same way or close to the same way.

[Xion X2]

there are no facts there thats the problem, you are presuming there is, but no with out standerization, you can't have theory as fact, it will always be theory. Its not even theory its guess work. If you follow the lines of logic you end up with too much variation in the test cases to make a good judgement. Did the others do the same tests you did, with the same number for frames, same time runnings? There is no way I can make that assumption.

variation is not only from your system, if you understood that then you would have your friends do the exact same tests as you. Thats why we called it fud over at rage, and its still fud without the other data, I already said it should end up close in this game anyways, variation from what becco and I did was there because we weren't able to do the race exactly the same way or close to the same way.

Nothing is ever going to be for certain, razor; you just have to take a leap of faith once in awhile. You are never going to have two perfect tests between two cards and a perfect benchmark run. There are just too many variables in this biz to ever get everything perfect. There could be a hiccup in an operating system; one system may not have a certain Windows update, one system may have a bad stick of RAM; one system may be running the game on a CD w/ fingerprints that causes a hiccup from a read error; one system may have a bad videocard; one system may have it's AA settings in the driver borked; one system may have a borked filtering option in the control panel; one system may have a bad driver uninstall-->reinstall..

And etc, etc, etc, etc, ad nauseum ad infinitum. Get the picture?

All one can do is bench several test samples (which I did) at completely different locations to get a general idea of the kind of performance one can expect.

And besides that, that's not even what we're talking about here. We're talking about your claim that my screenshots weren't valid because they were in different locations and that the framerate on Test Drive "vary widely" as you said. Well, HERE ( http://www.rage3d.com/board/showpost.php?p=1335010657&postcount=321) I have posted 6 different benchmarks, 12 races in total, in 3 different locations to see if there's any merit to your claim. And what we see is there isn't.. much. If 2.6fps is what you consider "vary widely," then I'd advise you to take another course or two on the English language and maybe a math class or two, since you like to so casually throw insults in the direction of my own math skills.

At least I can recognize that 2.6fps is similar and not a wide variation in performance.

Like I said, run home lil' doggie. No more frisbee for today.

[Razor1]

Nothing is ever going to be for certain, razor; you just have to take a leap of faith once in awhile. You are never going to have two perfect tests between two cards and a perfect benchmark run. There are just too many variables in this biz to ever get everything perfect. There could be a hiccup in an operating system; one system may not have a certain Windows update, one system may have a bad stick of RAM; one system may be running the game on a CD w/ fingerprints; one system may have a bad videocard; one system may have it's AA settings in the driver borked; one system may have a borked filtering option in the control panel; one system may have a bad driver uninstall-->reinstall..

And etc, etc, etc, etc, ad nauseum ad infinitum. Get the picture?

All one can do is bench several test samples (which I did) at completely different locations to get a general idea of the kind of performance one can expect.

And besides that, that's not even what we're talking about here. We're talking about your claim that my screenshots weren't valid because they were in different locations and that the framerate on Test Drive "vary widely" as you said. Well, here I have posted 6 different benchmarks, 12 races in total, in 3 different locations to see if there's any merit to your claim. And what we see is there isn't.. much. If 2.6fps is what you consider "vary widely," then I'd advise you to take another course or two on the English language and maybe a math class or two, since you like to so casually throw insults in the direction of my own math skills.

At least I can recognize that 2.6fps is similar and not a wide variation in performance.

Like I said, run home lil' doggie. No more frisbee for today.


Very good at being what I call a person who thinks he knows something but knows nothing at all. Why don't you break down the the first race test and tell me what the frames are and then do a small test, lets say you stayed in the first 50% of that race for 50% more time, and the last part of the race was the same as another tester's test. Well your logic, your test, your benchmark, just failed. I don't take leaps of faith, if you believe in such things join a seance group, faith has nothing to do with math they are quite opposite, if you want to have no one refute what you say, don't post at all, because what you say is BS. And now I'm going to say it, your problems with your g80 was BS you don't know how to build systems thats why you had so many problems with your g80, because you are biased as hell, and probably broke your own system on purpose. Well you know what you still won't bet me on future benchmarks from Crysis and UT3, come one take it up, I want your system so I can put it under my car and run it over and post it on youtube.

Hmm probably can teach my dog more about statistics then you would learn in a life time or two.

[Obie]

Can we please stay on topic here? This thread provides a lot of information and so far has done so without arguing. Xion and others have provided a lot of feedback on G80 and R600 and I would prefer that it's not polluted. I am sure that I am not the only person here that appreciates all the information here in this thread.

Edit: Great, there is an ignore list option here.

[Razor1]

Well information is good if it comes from a semi intellectual point of view not from a person who calls others names for having different view points, Xion a person who is very happy with his purchase which is great for him, and an avid ATi lover, almost got banned from Rage, why is that because he doesn't know when to shut up, and he can't put anything up without looking like a fool.

Its great you guys came together and did the benchmarks, I have followed this thread quite a bit, but its not good that you have a person that doesn't or isn't open minded to anything but his own results.

[Obie]

Is there anyone where that has the COJ with the 1.1.1.0 patch that can show us some results? :ORDER:

[Xion X2]

And now I'm going to say it, your problems with your g80 was BS you don't know how to build systems thats why you had so many problems with your g80, because you are biased as hell, and probably broke your own system on purpose. Well you know what you still won't bet me on future benchmarks from Crysis and UT3, come one take it up, I want your system so I can put it under my car and run it over and post it on youtube.


http://i14.tinypic.com/4q736kw.jpg

[Razor1]

good movie ;)


I'm not going to be ignored, Daniel!

This is exactly how you sound :)