Didn't someone recently publish a study showing that tri-SLI had the least microstutter? Even less than quad?
Nvidia BigK GK110 Kepler Speculation Thread
- Thread starter A1xLLcqAgt0qc2RyMz0y
- Start date
-
- Tags
- nvidia
Didn't someone recently publish a study showing that tri-SLI had the least microstutter? Even less than quad?
boxleitnerb
Regular
You mean the botched THG review 1-2 years ago? Not really. They ran into a CPU bottleneck with 3 GPUs, thus microstutter obviously decreased. They only provided one single diagram with frametimes, so there is basically no data that would support their claims.
1 at $1000 obviously as the costs for producing and distributing 2 are double that of 1. i.e. more profit.
2 at $600 or 1 at $1000 is another question though. But the crazy thing is even $600 is ridiculously expensive. $1000 is totally insane.
UniversalTruth
Veteran
These distributing costs... given that you can buy from china and ship it to Europe for free, then I would assume that these distributing costs are measured in pennies.
Actually, bulk container shipping is so cheap that it might as well be free. Takes a week or two, but it's very cost-effective.
1K €. Congratulations. I don't think they will sell a lot of these.
http://www.2compute.net/Asus/GTXTITAN-6GD5.htm
http://webshop.eurosys.be/en/compon...xtitan-6gb-graphics-card-gtxtitan6gd5-p387473
If I am not mistaken, these 1k prices include taxes (VAT).
If the Geforce Titan 6GB version is released with an MSRP anywhere close to $849 USD, then that will be a fair and reasonable price given that the Asus Ares 2 HD 7970 GHz Ed. 3GBx2 version sells for $1499 USD (http://www.newegg.com/Product/Product.aspx?Item=N82E16814121717&Tpk=Ares%202), the GTX 690 2GBx2 version sells for $999 USD (http://www.newegg.com/Product/Product.aspx?Item=N82E16814130781), the HD 7970 GHz. Ed. 6GB version sells for $599 USD (http://www.newegg.com/Product/Product.aspx?Item=N82E16814202005), and the GTX 680 4GB version sells for $529 USD (http://www.newegg.com/Product/Product.aspx?Item=N82E16814500288). Geforce Titan should have far higher performance per watt compared to all 7970 GHz Ed. and 7970 GHz Ed. x2 cards, should have somewhat higher performance per watt compared to all GTX 680 cards, and should have similar or better performance per dollar compared to the 6GB 7970 GHz Ed. and 4GB GTX 680 cards.
Last edited by a moderator:
Seemed far more recent. Think it had 79xx and Kepler in it... I'll dig for it in a bit.
Silent_Buddha
Legend
Of course, neither company mass produces cards in house in any appreciable quantity. I can't even remember which was the last product generation where AMD manufactured a significant number of cards themselves. It has been a long long time since they did that. The HD 2xxx generation? I can't remember if they were still making cards themselves then or if all the ATI branded cards were being made by Sapphire at that point.
Regards,
SB
I believe ATI's manufacturing of retail products ended with the All in Wonder.
boxleitnerb
Regular
Where does it say anything about 3-way SLI/CF?
Sorry. On vacay with only a phone and lots of family. Might not be the correct thread as I've not had time to dig through it.
UniversalTruth
Veteran
Speaking about bottlenecks... just look how poor the Intel Core i7-3960X Extreme Edition (3.30GHz) is. A single game but... nevertheless.
http://www.techspot.com/review/603-best-graphics-cards/page4.html
boxleitnerb
Regular
^ that is a matter of perspective. If your system is bogged down due to the CPU, no matter the exact reason, it's a bottleneck. If one were to follow your logic, you could blame almost every game out there for piss-poor coding because it doesn't run 6x as fast on y hexacore than on a singlecore.
Speaking about bottlenecks... just look how poor the Intel Core i7-3960X Extreme Edition (3.30GHz) is. A single game but... nevertheless.
http://www.techspot.com/review/603-best-graphics-cards/page4.html
I don't get what point you're trying to make? Clearly the Core i7-3960X isn't a poor CPU and that game is simply CPU heavy + likely doesn't make full use of all 12(!) threads. And it's still able to hit 75fps which should be enough by anyone's standards.
boxleitnerb
Regular
I think that is just semantics. A bottleneck is a bottleneck, the actual cause of this bottleneck doesn't change this. Btw I don't find it smart to judge games this way. We don't know how (in)efficient the code really is, we know nothing about the background. Maybe it just cannot work any better, who knows?
It would be equally silly to say that games are badly coded because there is a GPU bottleneck at 1600p with 4xAA. Bottlenecks are completely normal.
Similar threads
