Just bought a Sapphire hd 7990 video card

[bystander36]

This is really not representative of actual performance. For example BF3:

When 3x7970 scales, they destroy 2 Titans, by a decent margin. All you need is CF profiles updated. Anyways if I were the OP, I would not go tri-fire/tri-sli. I think with the HD7990 is enough.

Actually, the other one is more representative, because he is not using 5760x1080p resolution. He is at 1600p.

For some reason, 1 7990 isn't enough for him.

 

[bystander36]

It will help this explains it all its ashamed no one really post about hd7990 in crossfire yet damn it but here's a good one

http://www.kitguru.net/components/graphic-cards/zardon/sapphire-hd7990-quadfirex-review/

30 inch monitor

What happens on games which do not scale? In some cases you lose performance. It's your money, but you're more likely to enjoy 780 SLI over 7990 quadfire, unless benchmarks are all that matter to you.

The problems occur in other games:
http://tpucdn.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/images/alanwake_2560_1600.gif
http://tpucdn.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/images/ac3_2560_1600.gif
http://tpucdn.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/images/arkhamcity_2560_1600.gif

In those, and more, 3-way crossfire lowered performance.

 

[Rvenger]

Nice spec going sli 780? What monitor you using?

Not going SLI 780. I am quite happy with one and so is my wallet


I am using a Planar 27" 1080P monitor. I wanted to be able to max out everything at 1080p. I will move up to 2560x1440 sooner or later and this card should support it ok.



OP, enjoy your HD 7990. Don't let the Nay Sayers get in your way. AMD and Nvidia both make good cards, period, end of story.

 

[Final8ty]

What happens on games which do not scale? In some cases you lose performance. It's your money, but you're more likely to enjoy 780 SLI over 7990 quadfire, unless benchmarks are all that matter to you.

The problems occur in other games:
http://tpucdn.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/images/alanwake_2560_1600.gif
http://tpucdn.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/images/ac3_2560_1600.gif
http://tpucdn.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/images/arkhamcity_2560_1600.gif

In those, and more, 3-way crossfire lowered performance.

Then you just disable one GPU or make your own profile until one officially gets made.
The same things happens with SLI at times as well and people making there own profile with SLI compatibility bits.

Nothing is absolute, its about the majority of the time.

 

[bystander36]

Then you just disable one GPU or make your own profile until one officially gets made.
The same things happens with SLI at times as well and people making there own profile with SLI compatibility bits.

Nothing is absolute, its about the majority of the time.

I realize the same happens with SLI, though less often, but 4-way Crossfire/SLI makes it far more common.

It is his money though, so he can throw it around however he likes, it just doesn't make the best sense. Heck, 3-way 780's would be the same cost, and definitely faster.

 

[Final8ty]

I realize the same happens with SLI, though less often, but 4-way Crossfire/SLI makes it far more common.

It is his money though, so he can throw it around however he likes, it just doesn't make the best sense. Heck, 3-way 780's would be the same cost, and definitely faster.

Quad SLI with the 6 series has been far worse with scaling than Quad CF over all.
There is nothing to argue about, with Quadfire 7990 he can choose what he likes for each game 1,2,3 or 4 GPUs.It not for you to deicide for others what percentage of scaling or percentage of games work with 1,2,3 or 4 GPUs for it to be worth it for others.

You state your personal opinion and that's basically it in that regard.

No different than a console only player stating the benefits of PC gaming is not worth the hassle over the plug and play of the console, some people think its worth it and some don't.

 

[bystander36]

Quad SLI with the 6 series has been far worse with scaling than Quad CF over all.

I'm not sure about that: http://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/22.html

In the 18 game review, 3-way SLI continued to scale and 4-way scaled even more, where as 3-way CF gained only 1% on average (unless you got o 3 monitors and it still didn't scale as much as 3-way SLI). Of course there was negative scaling in a few games, which might have done better had they not be counted at all.

Perhaps you mean to say that 4-way SLI runs into vram limitations more, which would be the case with 690's.

 

[Final8ty]

I'm not sure about that: http://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_Titan_SLI/22.html

In the 18 game review, 3-way SLI continued to scale and 4-way scaled even more, where as 3-way CF gained only 1% on average. Of course there was negative scaling in a few games.

Perhaps you mean to say that 4-way SLI runs into vram limitations more, which would be the case with 690's.

Yes lets take that one review as gospel and nothing will ever change from it, drivers and profiles and the individualise setup will make no difference and all.

How many Tri fire or quadfire users have you asked about scaling on there rigs ?

 

[bystander36]

Yes lets take that one review as gospel and nothing will ever change from it, drivers and profiles and the individualise setup will make no difference and all.

How many Tri fire or quadfire users have you asked about scaling on there rigs ?

All I can do is use reviews. Are you saying that asking random people online is more accurate?

 

[toyota]

Topic creator: I purchase a 7990 and I'm having a good experience.

Nvidia fans: OMG MICROSTUTTERING! OMG CROSSFIRE WORTLESS!

sorry but its some of the AMD fans that cant accept reality. I recommend AMD cards all the time but the 7990 I will not. but hey if you are oblivious and cant see all the well documented issues then by all means enjoy your $1000 card that performs like a $450 card in reality most of the time.

 

[Final8ty]

All I can do is use reviews. Are you saying that asking random people online is more accurate?

The wider the pool of information the better as there is far to many factors to just pick reviews.

You cant beat first hand experience.

Using the reviewers settings again

Unigine Valley

2 x GTX 690s

2 x HD 7990s from the review

4 x Titans

The GTX 690s and Titans were bottlenecking where the reviewers settings are too low, 4.4ghz on the CPU is not enough.

http://forums.overclockers.co.uk/showthread.php?t=18516887

 

[Final8ty]

sorry but its some of the AMD fans that cant accept reality. I recommend AMD cards all the time but the 7990 I will not. but hey if you are oblivious and cant see all the well documented issues then by all means enjoy your $1000 card that performs like a $450 card in reality most of the time.

I wont recommend the 7990 either but i also wont recommend the 690 or titan also, but if that's what makes then happy then no one has a right to get in there way or put them down for doing so.

 

[bystander36]

The wider the pool of information the better as there is far to many factors to just pick reviews.

You cant beat first hand experience.

A wide pool is nice, but shouldn't you test more than benchmarks (a single benchmark at that)? These are gaming cards, not benchmarking cards. There are several benchmarks these days that definitely favor AMD, but actual games are more neutral.

EDIT: correct me if I'm wrong here, but the benchmarks you showed, have the 690 quad-SLI beating 7990 quadfire. That doesn't show what you are wanting to...though it appears to be a CPU bottleneck based on how titan quad-sli did the same as 690 quad-sli

 

[toyota]

I wont recommend the 7990 either but i also wont recommend the 690 or titan also, but if that's what makes then happy then no one has a right to get in there way or put them down for doing so.

we are talking about someone looking at top of the line $1000 cards not just cards in general.

and we have technical reviews for a reason as those will point out REAL issues that can be reproduced and documented. most users opinions are worthless because they dont have a clue what they are doing and only want to defend their purchases.

 

[bystander36]

we are talking about someone looking at top of the line $1000 cards not just cards in general.

and we have technical reviews for a reason as those will point out REAL issues that can be reproduced and documented. most users opinions are worthless because they dont have a clue what they are doing and only wont to defend their purchases.

Not only that, he is talking about adding a 2nd one. That is where I jumped in.

 

[Final8ty]

A wide pool is nice, but shouldn't you test more than benchmarks (a single benchmark at that)? These are gaming cards, not benchmarking cards. There are several benchmarks these days that definitely favor AMD, but actual games are more neutral.

You missed the point, the point was that user has the same cards and on the same tests and settings his show better performance.

 

[Final8ty]

we are talking about someone looking at top of the line $1000 cards not just cards in general.

and we have technical reviews for a reason as those will point out REAL issues that can be reproduced and documented. most users opinions are worthless because they dont have a clue what they are doing and only want to defend their purchases.

Right and your going to tell us which users or reviews we should all abide by i guess, reviews have been know to be wrong many times.

 

[bystander36]

You missed the point, the point was that user has the same cards and on the same tests and settings his show better performance.

Oddly, the list you showed has the 690 quad-sli doing better than 7990 quadfire...which I mentioned in my edit. I think a CPU bottleneck is holding them back, as the Titan quad-sli didn't improve over 690's.

 

[Final8ty]

Not only that, he is talking about adding a 2nd one. That is where I jumped in.

Amd you has stated your opinion, but my first hand experience differs for the last 6 years with quadfire over others opinions which is based on a snapshot of reviews and not on first hand experience.

 

[SirPauly]

It will help this explains it all its ashamed no one really post about hd7990 in crossfire yet damn it but here's a good one

http://www.kitguru.net/components/graphic-cards/zardon/sapphire-hd7990-quadfirex-review/

30 inch monitor

Thanks for that and there may be misconceptions that there is poor scaling with Quad! Quad makes sense for gamers that crave higher quality settings like SSAA and higher levels of transparency AA as well, imho.

 

[Final8ty]

Oddly, the list you showed has the 690 quad-sli doing better than 7990 quadfire...which I mentioned in my edit. I think a CPU bottleneck is holding them back, as the Titan quad-sli didn't improve over 690's.

Missed the point again :sigh:
Its my thread i know what has been said already.

 

[Final8ty]

Thanks for that and there may be misconceptions that there is poor scaling with Quad! Quad makes sense for gamers that crave higher quality settings like SSAA and higher levels of transparency AA as well, imho.

That,s what i use it for most of the time.

Its surprising how little GPU power you can get away with when your not trying to pile on quality and AA.

 

[bystander36]

Amd you has stated your opinion, but i first had experience difference for the last 6 years with quadfire or others opinions which is based on first had experience and not a snapshot of reviews.

The problem I have, is you have experienced it with one setup, I assume, since you still have 5970's. Reviews actually compare them side by side, so they can experience the difference.

I would think that someone with quadfire would still recommend going 2-way SLI/CF if they can get the same performance. It just simplifies things a bit.

 

[Final8ty]

The problem I have, is you have experienced it with one setup, I assume, since you still have 5970's. Reviews actually compare them side by side, so they can experience the difference.

I would think that someone with quadfire would still recommend going 2-way SLI/CF if they can get the same performance. It just simplifies things a bit.

The 5970's was not around 6 years ago so its not my first multi gpu setup, i had 2 more before then.
I also have 3 way 7950.

2 way has never been enough yet for me.

Its simple, quadfire and TriFire has been fantastic for me, i have also seen it being a nightmare for others, there is no guarantee how its going to be for another person so i don't encourage it but neither do i discourage it, if they want it then they can have it.

I have seen games go from a stutter fest even with Vsync to smooth as butter with a new driver, shite happens even on single gpus but people do not give up and go console because of it, its all about what matters most to the individual which some people seem to forget.

I state the benefits and the possible negatives but at the end of the day its up to them.

It works for me the majority of the time and that all that matters to me, at least i have the option of more power most of the time rather than not having it at all and needing it.

 

[bystander36]

You missed the point, the point was that user has the same cards and on the same tests and settings his show better performance.

(so I address your point)
I missed your point, because you didn't show an example of 2 or 3-way crossfire to compare to the 7990 quadfire.

That said, my point was that many times it doesn't scale, or it doesn't scale well. That benchmark generally favors 7970's over 680's. It should have shown the 7990x2 beating 690x2 if it scaled well, but in this case, it probably was a CPU bottleneck, so it is uncertain.