Originally Posted by Alatar
Originally Posted by SniperTeamTango
And you know nothing about marketing if you truly don't believe the answer doesn't lie somewhere in the middle.
Did Nvidia have a financial incentive to push FCAT (note, that's what they pushed, this frame metering stuff has been out for a while) before Titan's launch? Yes. Does that make the issue any less real? No.
But your entire argument in every thread is based on that people will pay a premium for even the smallest of gaps in gains, so why didn't AMD charge more for the 7970s? Oh right. market stabilization.
No, I just said that with the Titan you aren't paying a premium for the smallest gap ever. You're actually getting 32% extra over a 7970GHz which is a huge difference by any standards in the GPU world.
Remember building PCs has to be a better alternative than buying a console, loosing both companies customers doesn't benefit anyone. (Aka killing a market with inflation)
Not every part has the same purpose. Titan isn't there for the budget buyer. That's what the lower end cards are for.
Entirely your opinion but again you're welcome to it. Something completely impercieveable to me isn't worth an extra $700 (larger than the $600 gap between a 30k and 60x here btw, which I find interesting, finland, right?)
32% extra perf and better frame times isn't perceivable to you? Why not use a 7850 or something then? And the price gap between a 7970GHz and a Titan is just over 500€ here. Price gap between a 3930K and 3960X is around 450-500€ depending on the store. I've paid 1100€ because I've bought two.
Nothing wrong with doing it, alot wrong with condemning others for not doing it.
No one has condemned people for going the price/perf route. If anything there are a bunch of people saying that Titan users got scammed, that they cause high pricing or have more money than sense.
What percentage of this thread has been about frametime issues, price performance, etc, etc.
Don't see how this is relevant to my point. As I said Nvidia cards don't at the moment have as big frame time issues as AMD cards do. Since the thread is about 690 vs. titan it's obvious why frame times come up.
Forgive me for not taking a test entirely seriously when I do not personally get to see numeric results. Not to mention, there's 0 qualitative statement there. Its just a number of people.
It was a small test of 10 people, each one of them chose the 690 setup as the smoother one. You can find a video and some other stuff in Tom's 7990 review. If it's still invalid in your opinion then you're basically telling me that the test is a lie simply because they didn't make a table that says all users preferred the 690 setup.
Really so you're telling me your eye can track the path of a speeding bullet? Iiiiiinteresting. Let me guess you can also tell when a lightbulb turns on in the distance? And how long it takes the light to reach your eye too? We are talking about occasional issues, with milliseconds of issues.
Not understanding the issue at all which is probably why you're making all these strawman arguments. See the problem here is that you're saying we can't tell the difference when it's about just a few milliseconds.
If that's truly your opinion then let me ask you something; Can you tell the difference between 45 and 30fps? I'm sure most enthusiasts would tell you that they can. Then according to your logic they can see a ~10ms difference.
Unfortunately saying that you can't notice anything in x milliseconds is missing the point entirely.
And I can explain why that pill made entirely of sugar cured your headache too. Come on.
No, I'm saying aspirin cured my headache. Comparing the issues to a placebo is extremely invalid unless all the setups perform and act exactly the same.
Making me compare two Titan setups and saying one of them was a 7990, and then me saying that I see the differences would be a placebo effect. Seeing the difference between the 7990 and the Titan isn't because the difference is actually objectively there
You really mean to tell me occasional things like that are really, really, ruining your gaming experience? Something you definitely wouldn't get in benchmarking AFAIK, and honestly, you're playing the wrong games if that is where your focus is.
I'm playing the wrong games because I want smooth gameplay? And yes, gameplay that looks like 30fps while I'm actually getting 60 is ruining my gameplay experience.
And I can show you pictures of crops with interesting patters in them showing how they might be aliens trying to communicate with us all day long too, but would you take me seriously?
Really? I mean really? Crop circles? Comparing them to actual proof posted? Can we please have a conversation about this instead of just pulling complete strawmen from absolutely nothing.
Lab tests and actual perception are 2 completely different animals.
Not really when you see the same stuff in both situations.
It's fine to say you don't see the effect and if so then good for you. But just because you say that doesn't mean it doesn't exist. It's been proven to exist and has been proven to be very much noticeable by users.
Oh, and just a reminder, seeing as titan owners were screaming this across the boards when it came out, 7990 is not even in public hands yet (at least not in my country's) so let the drivers mature more. The arguments work both ways guys.
I don't argue that Titan drivers should mature. I think everything should work out of the box. The 7990 doesn't.
EDIT FOR FAIR ARGUMENT SAKE: I did watch the video until the slowmos just now.
Know what I think it is, I don't play games at these retardedly demanding settings and look for these things, I did see a couple (1-2) frames out of sync, and yes, online that would bother me. I do however refer to my driver maturity argument.
If you can't see the difference then I don't know what to tell you. That's the difference between 20 and 40 fps... So great for you I guess? You obviously can't see framerate differences or big stutters in gameplay that have been proven to be there.
Also, side note as I would like to learn, howcome these measuring tools are just hitting the floors now if this has been known about for a couple years?
Because the issue is somewhat complicated and software development takes a while...
Originally Posted by KyadCK
"Titan solves that problem"... just LOL. I can't reply any other way. Every review for the Titan that has Crysis 3 in the lineup puts the Titan at 45 FPS or lower... at only 2560x1600. Let alone the 3440x1920 that I run. The card would die. You're dreaming if you think Titan solves anything.
In cases where the Titan is under 45fps you're not going to get much better results with lower end dual cards either.
So Titan doesn't solve anything but 8% faster 7950 CFX that can't deliver smooth frame rates does?
It does not come close to Dual GPUs. Again, you're dreaming. Just scrolling through reviews it's anywhere from "only" 25% slower to 40% slower. If you think that's "close", you must think the 680 and Titan are close too, in which case why didn't you buy one of those?
And see this:
Yes, such massive differences especially when you consider that single GPUs don't need as much fps to feel as smooth as dual GPUs do due to frame time variances. If want to keep 60fps you'll nee much lower frame times/higher frame rates with dual GPUs to always keep the frame time under 16.7ms which is what a steady 60fps is
And ah, the dreaded smoothness claim. If only you could actually back this up, ever, without the assistance of a slow down video!
It was just backed up with a non slow-mo video....
And for further proof just see any frame metering review from the past year. Full of proof, even AMD themselves have admitted the issues. Calling them fake is pure denial, nothing more.
Ya know, I wasn't actually going to post this originally, but I feel that if you're going to make a big deal out of a small issue most can't even see...
You spent all your money on a ferrari. All the people who bought GTs and souped them up are kicking your ass, and all you can find to complain about is the smoothness of the ride. The only problem, is the race is taking place on an airport runway, and only a few can even tell the difference.
Remember, it's still only a ferrari. You are not the fastest guy on the road. Know where you stand alatar, because I notice how it's only the guys who don't have the GT who claims the ride sucks, and it's the ones who bought a ferrari who complain the loudest.
(Thinking you can even hope to catch those GTs just shows how bad you wish your ferrari was better too)
If there ever was a bad time for a car analogy it's this one.
If all you care is fps that doesn't correlate with the gaming experience then great. But please do not refute proven facts by simply saying that people don't notice them. Obviously they do since pretty much all reviewers noted the smoothness issues with their 7990 reviews.
I'll take a comfortable slower car as my daily driver any day. Especially when the only place where you can use that speed of yours is the runway. While in normal use you're limited to 60mph. See I can do it too.
Empty test benches are ugly too btw, they're only pretty when they look like this:
And/or (preferably and) have an LN2 pot on them.
Not really, managing 4 GPUs just messes up all the cable management.
I'll take the clean route:
Especially with extreme cooling that needs space: