G70 does 7703 in 3DMark05

[apoppin]

Originally posted by: Ronin
Did you ever stop to think that some of these cards were provided at lower than retail spec speeds? There's some food for thought for you.

Not that 2k5 is a representation of anything but a sythentic (and worthless, in my opinion) benchmark. Try using something that scales to your PC as a whole, such as Aquamark, and use THAT as a reference. If you want to use a crappy benchmark that can't even utilize what's offered in the 7800, more power to you, but you'll only end up disappointed.

actually i was the one that suggested there were several versions of the 7800 available for review . . the inq *confirmed* that there would be several varieties of 7800 offered for sale immediately after launch - part of nVidia's 'grand misdirection' [imo]. . . . however, YOU have been pretty negative on anyone's links or leaks while offering nothing in return.
:thumbsdown:

 

[Drayvn]

Originally posted by: apoppin

Originally posted by: Ronin
Did you ever stop to think that some of these cards were provided at lower than retail spec speeds? There's some food for thought for you.

Not that 2k5 is a representation of anything but a sythentic (and worthless, in my opinion) benchmark. Try using something that scales to your PC as a whole, such as Aquamark, and use THAT as a reference. If you want to use a crappy benchmark that can't even utilize what's offered in the 7800, more power to you, but you'll only end up disappointed.

actually i was the one that suggested there were several versions of the 7800 available for review . . the inq *confirmed* that there would be several varieties of 7800 offered for sale immediately after launch - part of nVidia's 'grand misdirection' [imo]. . . . however, YOU have been pretty negative on anyone's links or leaks while offering nothing in return.
:thumbsdown:

By the way, even if there is a higher spec card, dont you think it would be way way way overpriced at like $749 dont you think?

 

[Sentential]

Originally posted by: Ronin
Did you ever stop to think that some of these cards were provided at lower than retail spec speeds? There's some food for thought for you.

Dude they were clocked at 430 / 1200. Thats the reference clock. How much faster could they POSSIBLY scale it to. It cant be higher than 500 because that is what the current max appears to be.

Even if it IS 500 we already know what the scores will be like.

"Crappy Benchmark"? So wtf ARE we supposed to use then? Something that nV approves of? c'mon...

 

[Gamingphreek]

Originally posted by: Sentential

Originally posted by: Ronin
Did you ever stop to think that some of these cards were provided at lower than retail spec speeds? There's some food for thought for you.

Dude they were clocked at 430 / 1200. Thats the reference clock. How much faster could they POSSIBLY scale it to. It cant be higher than 500 because that is what the current max appears to be.

Even if it IS 500 we already know what the scores will be like.

"Crappy Benchmark"? So wtf ARE we supposed to use then? Something that nV approves of? c'mon...

3dMark 05 is a pure synthetic benchmark while Aquamark is an actual real game (Aquanox) therefore you get a much more accurate result.

Additionally from what i hear the 7800 is VERY CPU limited so far, even at 16x12 with 4x AA.

Dravyn... the MSRP is no where near $750. Just because something retails at that value to begin with doesn't mean that it will stay there. Additionally seeing as how the card isn't even released yet, that can be a very innacurate assumption.

Seriously people aren't you assuming a little too much a little too early. Wait to judge the card until we get some surefire legit reviews. Nothing against the other places, however leaking benchmarks while under NDA, sort of brings them down a notch in my book.

-Kevin

 

[ponyo]

Judging from OPPainter OC GF 7800 score, 7703 sounds about right. Looks like Inq was on the money.

 

[Drayvn]

Originally posted by: Gamingphreek

Originally posted by: Sentential

Originally posted by: Ronin
Did you ever stop to think that some of these cards were provided at lower than retail spec speeds? There's some food for thought for you.

Dude they were clocked at 430 / 1200. Thats the reference clock. How much faster could they POSSIBLY scale it to. It cant be higher than 500 because that is what the current max appears to be.

Even if it IS 500 we already know what the scores will be like.

"Crappy Benchmark"? So wtf ARE we supposed to use then? Something that nV approves of? c'mon...

3dMark 05 is a pure synthetic benchmark while Aquamark is an actual real game (Aquanox) therefore you get a much more accurate result.

Additionally from what i hear the 7800 is VERY CPU limited so far, even at 16x12 with 4x AA.

Dravyn... the MSRP is no where near $750. Just because something retails at that value to begin with doesn't mean that it will stay there. Additionally seeing as how the card isn't even released yet, that can be a very innacurate assumption.

Seriously people aren't you assuming a little too much a little too early. Wait to judge the card until we get some surefire legit reviews. Nothing against the other places, however leaking benchmarks while under NDA, sort of brings them down a notch in my book.

-Kevin

Very true, i should have waited to the NDA is lifted, but i was talking about eh MSRP of the rumoured Ultra not the GTX.

Im just saying the if the GTX is $649, any higher spec card that comes out after that would be higher than that. And if prices are higher than 650, then well thats a bit of a rip off aint it?

 

[Shamrock]

Dont forget, this isnt the "ultra" version, this is the "GT" version.

 

[Pete]

Ronin, two full reviews have shown the stock GTX to be 430/1200. Whether or not XFX or BFG offer higher-clocked versions seems irrelevant.

We all know 3DM's limitations, but the fact remains that The Inq appears to have been right--for current CPUs and that benchmark. The FX-57 seems to boost scores quite a bit, though (I read 8800--nice boost). I can't believe the two reviews leaked so far, HardSpell and PCOnline, didn't even bother benchmarking Far Cry or SC:CT HDR or Riddick SM2++ or anything more much more unusual than FC + D3 + HL2. Surely there are other (less publicized/optimized) games out there? And higher resolutions than 16x12?

Kevin, AM3 seems to be a better benchmark for AN3, not anything else. Just like benchmarking D3 is really mainly helpful for D3 performance.

Both of you seem to be protecting the GTX from what apears to be disappointing numbers. Anyone who just reads the spec sheets (50% more pixel pipes at a higher core clock) should know that, if nothing else (alpha AA, more capable shaders, etc.) the GTX will be a big improvement ... in games that can take advantage of it.

 

[apoppin]

Looks like the Inq was right again about the 7800GTX scores
[and they did so in a MATURE manner]

The Inq: Wins
:thumbsup:

Ronin: Loses
:thumbsdown:

i know who to pay attention to in the future
[i already knew]
:roll:

 

[zendari]

Originally posted by: apoppin
Looks like the Inq was right again about the &800GTX scores

The Inq: Wins
:thumbsup:

Ronin: Loses
:thumbsdown:

i know who to pay attention to in the future
[i already knew]
:roll:

QFT.

 

[Gamingphreek]

Let him defend himself first.

Maybe he has some explanation for his comments.

-Kevin

 

[dfloyd]

True maybe he does, maybe he was just really hoping they would be better and heck he had a fifty fifty chance of being right.... Right?

 

[KevinH]

Originally posted by: Gamingphreek

Originally posted by: Sentential

Originally posted by: Ronin
Did you ever stop to think that some of these cards were provided at lower than retail spec speeds? There's some food for thought for you.

Dude they were clocked at 430 / 1200. Thats the reference clock. How much faster could they POSSIBLY scale it to. It cant be higher than 500 because that is what the current max appears to be.

Even if it IS 500 we already know what the scores will be like.

"Crappy Benchmark"? So wtf ARE we supposed to use then? Something that nV approves of? c'mon...

3dMark 05 is a pure synthetic benchmark while Aquamark is an actual real game (Aquanox) therefore you get a much more accurate result.

Additionally from what i hear the 7800 is VERY CPU limited so far, even at 16x12 with 4x AA.

Dravyn... the MSRP is no where near $750. Just because something retails at that value to begin with doesn't mean that it will stay there. Additionally seeing as how the card isn't even released yet, that can be a very innacurate assumption.

Seriously people aren't you assuming a little too much a little too early. Wait to judge the card until we get some surefire legit reviews. Nothing against the other places, however leaking benchmarks while under NDA, sort of brings them down a notch in my book.

-Kevin

Gamingphreek - That's sort of missing the point on why everyone is taking an "I told you so" stance with Ronin. He was blasting every single leak including the aforementioned 3dmark 2005 scores as being bogus , fake, nowhere near the truth yadda yadda when in fact those scores were entirely on the mark. I dont think anyone will dispute the relevance of a 3dmark score with regards to actual game performance....however that's not the point of contention. The fact of the matter is, he said all those leaks were BS when in fact the they weren't.

 

[dfloyd]

Oh no the world has ended....

Link

Is a single Ati card actually beating the 6800 Ultra in Doom III?????

An Ati Card???

It cant be, just ask Rollo.

heh

Sorry I am in a wiley mood today.

Edit: What does this mean for Quake IV, Prey, Etc, Etc, Etc,

 

[Hajime]

Originally posted by: dfloyd
Oh no the world has ended....

Link

Is a single Ati card actually beating the 6800 Ultra in Doom III?????

An Ati Card???

It cant be, just ask Rollo.

heh

Sorry I am in a wiley mood today.

Edit: What does this mean for Quake IV, Prey, Etc, Etc, Etc,

I'd note that it only comes ahead in one test, and the article writer himself states that " Like we've seen in a few other games, though, the NVIDIA drivers don't seem to handle 2048x1536 very well. With AA/AF enabled, the 6800U once again takes a 50% performance hit when increasing the resolution". ]

 

[Ronin]

Inquirer loses. You lose. zendari is still a moron, and the 7800 validates pretty much all good claims (not the shvtty ones that came out of irreputable sources.

http://server.counter-strike.net/images/misc/2k5stock.png
http://server.counter-strike.net/images/misc/2k5stock2.png
http://server.counter-strike.net/images/misc/2k3stock.png
http://arc.aquamark3.com/arc/arc_view.php?run=959685271

What now, idiot? THESE are on the mark, thanks. Look at the dates as to when they were run, and note that none of the apps reflected proper core/memory speeds.

however that's not the point of contention. The fact of the matter is, he said all those leaks were BS when in fact the they weren't.

They were BS and then some, as you see above. Anything else?

 

[dfloyd]

Originally posted by: Hajime

Originally posted by: dfloyd
Oh no the world has ended....

Link

Is a single Ati card actually beating the 6800 Ultra in Doom III?????

An Ati Card???

It cant be, just ask Rollo.

heh

Sorry I am in a wiley mood today.

Edit: What does this mean for Quake IV, Prey, Etc, Etc, Etc,

I'd note that it only comes ahead in one test, and the article writer himself states that " Like we've seen in a few other games, though, the NVIDIA drivers don't seem to handle 2048x1536 very well. With AA/AF enabled, the 6800U once again takes a 50% performance hit when increasing the resolution". ]

Sorry friend you have to understand where I am coming from. In almost every video card comparison on this forum that I have seen or read Rollo has went out of his way to point out that Ati cards are not up to par when it comes to Doom III (Or any future game based on the engine) when compared to Nvidias 6800 series cards. Considering the fact that he constantly points this out, I find that its only fair to return the favor and point out that he was wrong. According to those benchmarks not only can Atis cards do pretty well at Doom III (And for Rollos comparisons, Prey, Quake IV, and such as well) but it can even beat what until today was Nvidias top end single card solution. And it does this in the game that has been the strongest argument for diehard Nvidia fans.

Now it is also supposedly true that the main reason Ati is weaker at OpenGL is because of their weaker OpenGL drivers. (This would make sense considering Ati cards, even down to the 9800 Pro do very well in most games, so obviously the hardware is powerful (In fact the 9800 Pro is faster than the 6600 GT in several games, considering the 6600GT is a full generation beyond the 9800 Pro, that is saying something). So takeing into account that Ati can beat the 6800U in Doom III with their openGL drivers as they are proves that Rollo has been wrong on every occasion where he claimed that the 6800 based Nvidia cards were far superior on the Doom III engine when compared to Ati.

And also friend I am not sure what you are saying above. By your statement that the reviwer himself writes " Like we've seen in a few other games, though, the NVIDIA drivers don't seem to handle 2048x1536 very well" I dont gather what you are saying. From what I read of that he is saying the 6800U is weaker than its Ati counterpart. Looking at the numbers at the highest res, I would tend to agree.

Also I count two tests where the Ati surpasses the 6800 Ultra. With all this said though I hope everyone can gather one thing from this. Atis cards are not as bad as people claim in Doom III based games. If anything its at least matching the 6800 Ultra within a few fps and bettering it at the max res with AA/AF on.

The fact I find most impressive of all is that the Ati card is within 8.2 FPS of the 7800 GTX at Max Res with AA/AF on. Considering that if I were Nvidia I would be worried about whats just around the corner.

 

[apoppin]

Originally posted by: Ronin
Inquirer loses. You lose. zendari is still a moron, and the 7800 validates pretty much all good claims (not the shvtty ones that came out of irreputable sources.

http://server.counter-strike.net/images/misc/2k5stock.png
http://server.counter-strike.net/images/misc/2k5stock2.png
http://server.counter-strike.net/images/misc/2k3stock.png
http://arc.aquamark3.com/arc/arc_view.php?run=959685271

What now, idiot? THESE are on the mark, thanks. Look at the dates as to when they were run, and note that none of the apps reflected proper core/memory speeds.

however that's not the point of contention. The fact of the matter is, he said all those leaks were BS when in fact the they weren't.

They were BS and then some, as you see above. Anything else?

Yes, what are you talking about?

the Inq make ONE claim for the 7800GTX: ~7700 marks in 3DMark05 . . . several sources HAVE confirmed it. You said it was BS and WAY off.

Even "your" +600 is not "way off" . . . only about 7%


i'd really like an explanation - withOUT flaming - please.

 

[Creig]

Inquirer report that according to Hardspell, G70 does 7,737 in 3DMark05.

Originally posted by: Ronin
Inquirer loses. You lose. zendari is still a moron, and the 7800 validates pretty much all good claims (not the shvtty ones that came out of irreputable sources.

http://server.counter-strike.net/images/misc/2k5stock.png
http://server.counter-strike.net/images/misc/2k5stock2.png
http://server.counter-strike.net/images/misc/2k3stock.png
http://arc.aquamark3.com/arc/arc_view.php?run=959685271

What now, idiot? THESE are on the mark, thanks. Look at the dates as to when they were run, and note that none of the apps reflected proper core/memory speeds.

however that's not the point of contention. The fact of the matter is, he said all those leaks were BS when in fact the they weren't.

They were BS and then some, as you see above. Anything else?

Sooo.....


I guess t-break.com are lying, too.

Athlon64 3500+
3DMark05 - 7,608



and Hot Hardware...

AMD Athlon 64 FX-55 (2.6GHz)
3DMark05 - 7,777



and Guru3D...

AMD Athlon 64 4000+, Socket939
3DMarks - 7,741



and Hexus.net...

AMD Athlon 64 FX - 2800MHz, 1MiB L2
3DMarks - 7,736



and PC Perspective...

Athlon 64 FX-55
3DMarks 7,741


and Bjorn3D...

AMD Athlon 64 FX-55
3DMarks - 7,641



and The Tech Report...

Athlon 64 4000+ 2.4GHz
3DMarks - 7,607



and Tom's Hardware Guide...

AMD FX-55 (2.6GHz)
3DMarks - 7,788



Must just be some massive conspiracy to discredit both you and Nvidia, Ronin.

 

[Insomniak]

I frankly don't give a damn about 3dmark05.


The two of you are now arguing over nothing except who was right and who was wrong. This has become an argument for ego. You're worried about your e-cocks.


3dmark, as a whole, has become completely irrelevant after 2k1, as it no longer uses genuine game engines. The facts about the GeForce 7800GTX:

1) 3dmark performance is mildly improved
2) Real world performance is amazing, and looks to be cpu limited at anything below 1600x1200 4xAA/8xAF.

Conclusion:

Screw 3dmark, screw this argument, and buy a 7800GTX if you want the best performance out of your GAMES, not your benchmarks.

 

[qbackin]

Originally posted by: Insomniak
I frankly don't give a damn about 3dmark05.


The two of you are now arguing over nothing except who was right and who was wrong. This has become an argument for ego. You're worried about your e-cocks.


3dmark, as a whole, has become completely irrelevant after 2k1, as it no longer uses genuine game engines. The facts about the GeForce 7800GTX:

1) 3dmark performance is mildly improved
2) Real world performance is amazing, and looks to be cpu limited at anything below 1600x1200 4xAA/8xAF.

Conclusion:

Screw 3dmark, screw this argument, and buy a 7800GTX if you want the best performance out of your GAMES, not your benchmarks.

Dam tell it like it is!

 

[Creig]

Originally posted by: Insomniak
I frankly don't give a damn about 3dmark05.

Then why are you posting in a thread titled "G70 does 7703 in 3DMark05"?

 

[Gamingphreek]

Originally posted by: qbackin

Originally posted by: Insomniak
I frankly don't give a damn about 3dmark05.


The two of you are now arguing over nothing except who was right and who was wrong. This has become an argument for ego. You're worried about your e-cocks.


3dmark, as a whole, has become completely irrelevant after 2k1, as it no longer uses genuine game engines. The facts about the GeForce 7800GTX:

1) 3dmark performance is mildly improved
2) Real world performance is amazing, and looks to be cpu limited at anything below 1600x1200 4xAA/8xAF.

Conclusion:

Screw 3dmark, screw this argument, and buy a 7800GTX if you want the best performance out of your GAMES, not your benchmarks.

Dam tell it like it is!

Get em Insomniak. Really though, i agree with you 100%.

 

[Insomniak]

Originally posted by: Creig

Originally posted by: Insomniak
I frankly don't give a damn about 3dmark05.

Then why are you posting in a thread titled "G70 does 7703 in 3DMark05"?

Because I'm attempting to get some of the minds that seem mired in here (Ronin et al) to turn their analytical lobes to something more productive than Pen0rMark05.

 

[jpeyton]

Originally posted by: Creig
Must just be some massive conspiracy to discredit both you and Nvidia, Ronin.

pWned!@ Thank you.