• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

GTX 780 Ti Specifications Leaked – Full Blown Gk110 Core with 2880 SP

melmac said:
Boom, i know you love Nvidia, but, come on, He has told some pretty good Lies before. Jokester made reference to one whopper he told. The two examples you gave for power use are bad examples. The 5800k to 6800k are APU's, the powersaving was made by managing the way the cpu and gpu work together better. For your other example, the 4870 and 4890, you do realise that the 4890 was a different chip entirely? RV770 to RV790. Even if they were the same chip, it still wouldn't support your argument, because while the 4890 used less power at idle, it did use way more under load.

We know the 780Ti is a GK110 chip. We know the GK110 is very efficient as it was designed to work in supercomputers with strict power demands. You think they are somehow magically create more performance for less power?

It will definitely be cooler and quieter than the 290x but will it be cooler and quieter than a stock 780?

But you would have to be living in dreamland to think it is going to use less power than the titan.
Click to expand...

Not really the 5800K > 6800K was an example everyone can follow, same chip, same die node, tweaked design for less power consumption and better overclocking. Same logic applies to GPU's. The latest GK110 revision, could have better power consumption to performance ratio.

Also the 4890 was exactly a tweaked 4870, they added a de-coupling ring, an additional 3million more transistors than 4870. Included improvements / tweaks made, but there is no logic design changes. Same architecture, same die node, resulted in better performance at stock and much better overclocking.

Likewise the GTX 780ti has more shaders, along with tweaked design. We just have to wait for reviews now to see how it compares.
 
Watch this...
Ok with a weak CPU and too much wattage.

attachment.php



attachment.php
 
Last edited:
Boomstick777 said:
Not really the 5800K > 6800K was an example everyone can follow, same chip, same die node, tweaked design for less power consumption and better overclocking. Same logic applies to GPU's. The latest GK110 revision, could have better power consumption to performance ratio.

Also the 4890 was exactly a tweaked 4870, they added a de-coupling ring, an additional 3million more transistors than 4870. Included improvements / tweaks made, but there is no logic design changes. Same architecture, same die node, resulted in better performance at stock and much better overclocking.

Likewise the GTX 780ti has more shaders, along with tweaked design. We just have to wait for reviews now to see how it compares.
Click to expand...

No Boom, sorry, the logic might apply to GPU's but not the way you are doing it. The 6800k is the second generation of the 5800k.

The 4890 was the second generation of the 4870. It was a different chip, lots of similarities but different, just like the 480 and 580, first and second generation fermi. But, you missed the point, or proved my point, yes it perform faster at stock and did perform overclock better, but it also used more power to do that.

Now, both the Titan and the 780 are second generation Kepler. We know the 780ti is also gk110. Its not a new chip like the 4890 was. They have done all the optimisations that they can do. The only way for them to increase performance is to increase clocks, unlock the cores and maybe change to faster memory. All of which require more power.

Which leads me to another question, you obviously believe all the leaked benchmarks that you have been posting, so how can believe them and not the power consumption figures with said benchmarks?
 
In my opinion if people are worried about power consumption they shouldn't be even looking at the top end cards. Anyone purchasing one of these cards should have higher end 750W+ power supplies and cases with good airflow. As long as they are within 100W of each other under extreme load it's well within the cooling and power capabilities of such setups.
 
Last edited:
melmac said:
Which leads me to another question, you obviously believe all the leaked benchmarks that you have been posting, so how can believe them and not the power consumption figures with said benchmarks?
Click to expand...

It's called confirmation bias. Basically absorb everything as absolute fact if it suits your agenda/beliefs/dreams/hopes and any info that does the opposite gets defelcted by the shield of fundamentalism.

http://en.wikipedia.org/wiki/Confirmation_bias
 
Gregster said:
Not sure why people are having a go at Boom. I say lay off him and his analogy does work.
Click to expand...

It's not about the analogy, it the fact that he takes the charts leaked so far as Gospel for performance but he chooses to ignore the power consumption on the same charts.

I'm not saying the charts are correct wither way, but if anyone is going to take them as facts they shouldn't cherry pick the parts they like and refute the parts they don't.
 
Last edited:
Overclocked 780 Ti Significantly Better than the R9 290X Read more: http://wccftech.com/overclocked-780-ti-benchmarks-l

Before we brought you Nvidia 780 Ti’s benchmarks, now we bring you its overclocked benches, only a few are available but believe me when i say, they do not disappoint. The source is again the credible Chiphell forums.
Overclocked 780 Ti Significantly Better than the R9 290X
Firs things first, the comparison is between an overclocked GTX 780 Ti with its stock self and the R9 290X. We would like to point out that unlike the R9 290X the 780 Ti’s cooler is quite sufficient for OC whereas the R9 290X is not. It might be more fair to wait for the Custom models of the R9 290X to show up before pitching in. Still the results speak for themselves.

Read more: http://wccftech.com/overclocked-780...massive-overclocking-potential/#ixzz2jfR0616H
Click to expand...
 
Well not read the last few pages on here yet, but just took a quick glance at videocardz.com...Those benchmarks!


And power consumption? Seriously having these debates still? The 290X isn't exactly eco friendly so why does it matter if the Ti consumes more? Or are we going to steer off course into the central heating gags.
 
Last edited:

So if the ti is clocked a little higher than the titan @ stock and is only giving out around 4-5% more performance which we can take from these leaked scores it's really not an upgrade to pretty much anyone on a Titan and those with high clocking gtx780's. Clock for clock with Titan it look's to only be about a 3-4% better if it is indeed clocked higher at stock.
 
6192 is the same as what WhyScotty scored and this 780Ti is reading over 1320Mhz. I call fake on the slides, purely on knowing what is what with Titans. I think that is a Titan mincing about as a 780Ti.
 
Gregster said:
6192 is the same as what WhyScotty scored and this 780Ti is reading over 1320Mhz. I call fake on the slides, purely on knowing what is what with Titans. I think that is a Titan mincing about as a 780Ti.
Click to expand...

Was thinking the same myself but i looked at the physics score (8283) on the benchmarks and it seems these scores were done with a low end cpu. With a decent cpu the score would be pretty good but probably not as good as Scotty's.
 
You must log in or register to reply here.
Back
Top Bottom