OP
- Joined
- Jan 2, 2012
- Thread Starter
- #181
Do we not have one?
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
FEATURED Nvidia 1080 launch thread
- Thread starter Soulcatcher668
- Start date
- Joined
- Jun 28, 2012
Should we trust these 3DMark results ? my Strix OC is faster then 1080 stock if we go by them
Can you reference one specifically instead of "these 3DMark results"?
- Joined
- May 29, 2005
Should we trust these 3DMark results ? my Strix OC is faster then 1080 stock if we go by them
can you please link to the exact page you are referring to?
And are you sure you are comparing same resolution/settings?
- Joined
- Aug 14, 2014
Am i reading them wrong ? should i compare total or just graphical result ?
http://www.gamespot.com/articles/nvidia-geforce-gtx-1080-review/1100-6439863/
http://www.guru3d.com/articles_pages/nvidia_geforce_gtx_1080_review,28.html
http://hexus.net/tech/reviews/graph...gtx-1080-founders-edition-16nm-pascal/?page=6
http://www.tweaktown.com/reviews/7703/nvidia-geforce-gtx-1080-review-hail-king-baby/index6.html
http://www.gamespot.com/articles/nvidia-geforce-gtx-1080-review/1100-6439863/
http://www.guru3d.com/articles_pages/nvidia_geforce_gtx_1080_review,28.html
http://hexus.net/tech/reviews/graph...gtx-1080-founders-edition-16nm-pascal/?page=6
http://www.tweaktown.com/reviews/7703/nvidia-geforce-gtx-1080-review-hail-king-baby/index6.html
- Joined
- May 29, 2005
I would only compare graphics results, as every other piece test system could be different (X79/X99/Z170/Z97/etc)
I don't know how far your card is overclocked. I imagine if you compared stock vs stock (as reviews do) you would see closer to the numbers that are given there.
If you have a solid overclock on your system it is quite possible to take a large chunk of the 20% improvement away from the 1080, ignoring what it is capable of when it is also overclocked increasing its scores by ~10%+
Example:
http://hexus.net/tech/reviews/graph...tx-1080-founders-edition-16nm-pascal/?page=12
5455 graphics score at 4K is 10% larger (rounding) than your 980ti overclocked.
I don't know how far your card is overclocked. I imagine if you compared stock vs stock (as reviews do) you would see closer to the numbers that are given there.
If you have a solid overclock on your system it is quite possible to take a large chunk of the 20% improvement away from the 1080, ignoring what it is capable of when it is also overclocked increasing its scores by ~10%+
Example:
http://hexus.net/tech/reviews/graph...tx-1080-founders-edition-16nm-pascal/?page=12
5455 graphics score at 4K is 10% larger (rounding) than your 980ti overclocked.
- Joined
- Aug 14, 2014
Sadly my Strix only does 1469mhz, so i could be classed as a bottom feeder
That's what i meant, i assumed the websites only gave out graphic scores instead of overall, as it is a GPU review. If that is the case only Hexus actually beats which is why i posted in the 1st place. If its overall score assuming stock CPU then sign me up for the next 1080 batch (or maybe i can wait for the Ti version) 
Why would i care about validation ? this was done with the latest 3dmark version, and i merely asked a question of the reviews validity against my own setup.
- Joined
- May 29, 2005
Sadly my Strix only does 1469mhz, so i could be classed as a bottom feeder
From what I can tell most of the sites gave graphics scores. Some may give combined, but that's overall a more useless number to give in many circumstances. So yes, your 980ti that is overclocked comes close/matches a 1080 at stock, but from looking at overclocked benchmarks (of which there aren't a ton yet) the 1080 looks to beat your score(s).
The reason to care about validation is that it lets everyone else looking at your scores know that you utilized approved graphics settings for the application (and also drivers, sometimes). A lot of the time 'non-valid' scores have various settings tweaked from defaults which causes them to be invalidated by futuremark. It's not that we don't trust your scores, but it adds a bit of validity to your screenshots.
- Joined
- Aug 14, 2014
Never noticed you could change the settings, i just open the program and press run
Even now i don't think theres any reason to change them, it should be ran at standard for a proper comparison right ? ill think about validation if i ever fork out the cash to buy the program...Any word on how the drivers are working ?
- - - Updated - - -
Aren't those peak temps with overclock ?
- Joined
- Jun 28, 2012
Not only does it validate benchmark settings and verify drivers, but it also allows there to be a website link to be shared.
It also makes sure the newest version of SystemInfo is running and there is no "timer is off" error occurring.
Until it can be validated, the benchmark is only useful for comparing to your own previous runs.
Also, remember, GPU Boost at the top end is a function of temperature threshold as well.
You're comparing a 3rd party cooling solution to a reference cooling solution.
Coolers like yours will keep the 1080 cooler as well, allowing for more effectiveness from GPU Boost.
It also makes sure the newest version of SystemInfo is running and there is no "timer is off" error occurring.
Until it can be validated, the benchmark is only useful for comparing to your own previous runs.
Also, remember, GPU Boost at the top end is a function of temperature threshold as well.
You're comparing a 3rd party cooling solution to a reference cooling solution.
Coolers like yours will keep the 1080 cooler as well, allowing for more effectiveness from GPU Boost.
- Joined
- Jun 6, 2002
maybe i missed one but i didnt see any numbers comparing to TITAN X. Not only that but im only seeing some numbers show the 1080 to be 25%-33%(give or take) faster in some cases vs 980TI. i was expecting more, maybe its driver related or something but with NV touting the 1080 like this. i would have expected better driver work for the 1080 even on the first release, maybe all that work just went into the new DOOM to show things off.
*shurgs*
*shurgs*
- Joined
- Jun 28, 2012
Basically the same as the 980Ti, like it has been since launch.
http://www.pcgamer.com/gtx-1080-review/
Looks to be 30-50% faster than the Titan X at 4K resolution based on the link above, pending which game you're looking at.
Why is everyone "expecting more" from this card...
its ~31% faster than 980ti, its significantly more power efficient, can apparently overclock to the moon, has new and updated feature sets that are actually USEFUL, extended VR support,
oh yeah. and its CHEAPER.
Seriously, what exactly where people expecting here???
its ~31% faster than 980ti, its significantly more power efficient, can apparently overclock to the moon, has new and updated feature sets that are actually USEFUL, extended VR support,
oh yeah. and its CHEAPER.
Seriously, what exactly where people expecting here???
- Joined
- Jun 28, 2012
Seriously, what exactly where people expecting here???
Santa?
I don't know, but I'm personally stoked to get one of these in my hands. It'll do wonders for 5760x1080.
- Joined
- May 29, 2005
Not sure exactly what else you were expecting from the card. It is an improvement across the board with normal games and has (for the most part that I understand) a huge improvement in VR.
- Joined
- Jun 6, 2002
i wasnt expecting anything other then what NV said it would do. i still dont see this twice as fast as Titan X so its my fault this card doesn't live up to what they said. GOT IT. maybe people have shorter term memory then i thought... my whole point about that post is about what NV claimed it would do....
