NVIDIA GeForce GTX 1070 Ti Founders Edition 8 GB

[Rahmat Sofyan]

Final question, so this is the last Pascal card ?

Or there will be more surprise from nVidia ?

 

[looniam]

no, NV has one more card up their sleeve; a slightly cut GP102 from the 1080Ti w/9Gbs of vram.

they're going to call it the 1080TiTs and start supporting SLI again. the marketing will be:

"everyone needs a pair of TiTs in their rig"

 

[NC37]

Well if this lowers the insane price on the Vegas, great. Unfortunately it won't really matter because of miners.

 

[StrayKAT]

Well if this lowers the insane price on the Vegas, great. Unfortunately it won't really matter because of miners.

I think it's made precisely to not give AMD any breathing room. Not even that little window between 1070 and 80.

 

[mastershake575]

I like how it took AMD 14 months to counter the 1070 and then Nvidia re-counters it in only two months time

I also like how they have third-party cooler options at launch. I have yet to see a Vega 56 third-party cooler model in the US even though it's been out two months......

 

[fullinfusion]

What was that Linus?.... Nooo, Just a mustache and wig?




Nvidia

I think he was right

@mastershake575 no supply and maybe?


Nvidia is just ... well, you know.... As said above

 

[Tsukiyomi91]

Decent card overall. But the lack of OCing it is what puts most of us off.

 

[fullinfusion]

@W1zzard so what do you actually set ROTTR at for AA and AF because I can set a bunch of different settings and get different numbers?

 

[W1zzard]

@W1zzard so what do you actually set ROTTR at for AA and AF because I can set a bunch of different settings and get different numbers?

I don't use the ingame benchmark

 

[fullinfusion]

I don't use the ingame benchmark

Ummm, excuse me then how do you do your test? s

I always assumed you did an in game bench so I've been miss leaded all this time is what your sayin?

DO you just pick a spot to record frames and that's it?

 

[W1zzard]

DO you just pick a spot to record frames and that's it?

Yes the ROTTR integrated benchmark isn't representative of typical game performance in my opinion.

"just pick" means play through the full game, observe performance, remember where good scenes/locations can be found for testing and then use one of them.

 

[TXST Guardian]

I don't get why you prefer Vega over this. it's much more efficient so it can run cooler and quieter than Vega and with overclocking it can be faster than both OCed Vega 56 and OCed Vega 64.

IF you could overclock it high enough. Isn't Nvidia limiting overclocking so it does not surpass the 1080? if so why would you want a sub 1080 card? That's why I said 1080 or vega

 

[W1zzard]

Isn't Nvidia limiting overclocking so it does not surpass the 1080?

Manual overclocking is not limited in any way. Out of the box overclocks aren't allowed.

 

[fullinfusion]

Yes the ROTTR integrated benchmark isn't representative of typical game performance in my opinion.

"just pick" means play through the full game, observe performance, remember where good scenes/locations can be found for testing and then use one of them.

well that isn't a solid base line imho, the bench mark is a solid baseline that cant be messed with or choosen what to base a performance number from..

In all honesty that's not fair because I can also pick and choose what shows better numbers for a given card and post it.. shame on you wizzard ...no disrespect but shame shame

 

[W1zzard]

because I can also pick and choose what shows better numbers for a given card

That's exactly why in-game benchmarks are bad. Developers are influenced by GPU vendors in terms of scene selection, scene duration (think temperature throttling), and GPU vendors heavily optimize their drivers just for the benchmark alone. Do you want review results showing how the actual game performs or how the benchmark performs?

Why would I pick a scene with better numbers for a given card?

 

[fullinfusion]

Don't know, and is why I mentioned it.

I've always thought most reviews for cards in here are based off a benchmark of such since it's available and less time consuming.

I play ROTTR all the time and some parts of the game don't hit 140 frames but other areas do. That's why I kinda Bustin your balls on this one just because how do we know what part of the game you pulling your numbers from.. that's all I'm getting at.

I spent last night running multiple settings to remotely get your numbers and I'm like wth is up with this and is why I asked what settings in AA and af are you using because I'm using the same test system as you but I'm using different memory and that's the only difference and scoring better numbers then the review? That's what confused me.