Vega/Navi Rumors (Updated)

[Despoiler]

Why do they run with v-sync on?

Being it's a preview on an very immature platform they don't want to give away what would seem like a full performance picture right now. It wouldn't be accurate.

It appears that Vega is close to Titan X. Interesting how mature are the drivers.

They aren't. As of about two weeks ago they were using slightly modified Fury drivers.

 

[Qwertilot]

Being it's a preview on an very immature platform they don't want to give away what would seem like a full performance picture right now. It wouldn't be accurate.

Significantly doubt marketing people (anywhere that is!) care about pedantic accuracy Not giving information away yes, but you also have to presume any sort of demonstration like this is at least mildly rigged to look good.

 

[swilli89]

It appears that Vega is close to Titan X. Interesting how mature are the drivers.

Very promising indeed! I hate that we will most likely have to wait until May for retail release, though. Perhaps the smaller Vega will release sooner and big Vega later? Similar to nVidia's rollout strategy.

 

[Despoiler]

Significantly doubt marketing people (anywhere that is!) care about pedantic accuracy Not giving information away yes, but you also have to presume any sort of demonstration like this is at least mildly rigged to look good.

There is definitely a marketing strategy being invoked. Tesla uses the same strategy. Trickle information on as of yet unreleased product to keep your company and product in the consumer mind. However, it's also true that AMD only got their first Vega silicon back in November/December. These are the very first boards built and are working on rudimentary drivers.

 

[Headfoot]

However, it's also true that AMD only got their first Vega silicon back in November/December. These are the very first boards built and are working on rudimentary drivers.

Source?

 

[Despoiler]

Source?

You should probably stay more current on news. All of this has been discussed on Anandtech and several other forums and/or tech press. AMD sent Vega off to manufacturing in late June. That was the Vega party Raja tweeted about. A new Fury GPU hardware ID showed up in the drivers in late October/early November. An AMD rep responded on Reddit that the Vega demo during New Horizons was using a slightly modified Fury drivers. Basically just enough modification to get everything to run.

 

[DeeJayBump]

...An AMD rep responded on Reddit that the Vega demo during New Horizons was using a slightly modified Fury. Basically just enough modification to get everything to run.

For clarity's sake: the Vega New Horizons demo used a Vega GPU running on modified Fury drivers, according to AMD.

 

[Bacon1]

You should check what map they are playing. That benchmark is from Tatooine which is much less GPU heavy.

http://www.legitreviews.com/nvidia-geforce-gtx-1080-founders-edition-video-card-review_181298/8

We chose the Ewok planet of Endor (Tana) for testing as it was more graphically demanding than scenes that took place on Hoth or Tatooine, for example. We also found the ‘Chase’ mission to be more demanding than the battle on Endor.

http://www.techspot.com/review/1096-star-wars-battlefront-benchmarks/

Since we know techspot benchs on Endor we can compare it and we know it will be closer than that from legit reviews which tests Tatooine:

http://www.techspot.com/articles-info/1246/bench/Battlefront_02.png

http://www.techspot.com/review/1246-ludicrous-graphics-test/

 

[ShintaiDK]

You should check what map they are playing. That benchmark is from Tatooine which is much less GPU heavy.

http://www.legitreviews.com/nvidia-geforce-gtx-1080-founders-edition-video-card-review_181298/8



http://www.techspot.com/review/1096-star-wars-battlefront-benchmarks/

Since we know techspot benchs on Endor we can compare it and we know it will be closer than that from legit reviews which tests Tatooine:

http://www.techspot.com/articles-info/1246/bench/Battlefront_02.png

http://www.techspot.com/review/1246-ludicrous-graphics-test/

When there is no ingame benchmark, how do you compare it.

I didn't see any demanding scene as such in the Vega run. It certainly wasn't the speeder run. And the faster paced the more GPU load.

And who knows how much the game have been optimized since.

 

[Bacon1]

When there is no ingame benchmark, how do you compare it.

I didn't see any demanding scene as such in the Vega run. It certainly wasn't the speeder run. And the faster paced the more GPU load.

And who knows how much the game have been optimized since.

Then why did you post benchmarks from May 18th to try to compare them in your post? At least the ones I referenced are on the same planet and within the last few months (September 26th)

 

[zinfamous]

When there is no ingame benchmark, how do you compare it.

I didn't see any demanding scene as such in the Vega run. It certainly wasn't the speeder run. And the faster paced the more GPU load.

And who knows how much the game have been optimized since.

so...you are now arguing that benchmarks you posted are irrelevant or, at best, incomparable?

 

[ShintaiDK]

Then why did you post benchmarks from May 18th to try to compare them in your post? At least the ones I referenced are on the same planet and within the last few months (September 26th)

so...you are now arguing that benchmarks you posted are irrelevant or, at best, incomparable?

Maybe you two should look up who started to post benchmarks and claim a certain performance level.

 

[Bacon1]

Maybe you two should look up who started to post benchmarks and claim a certain performance level.

So what was the point of your post with benchmarks exactly then? To show invalid data purposefully? You left off source links to make it harder to find the date and location of testing done.

 

[HurleyBird]

55 degree FOV is not standard, right? It looks like AMD cherry picked the exact situation where they could maintain ~60 FPS minimum on ultra, which is pointing to a card below Titan XP, assuming drivers aren't able to close the gap.

 

[zinfamous]

Maybe you two should look up who started to post benchmarks and claim a certain performance level.

You challenged those benchmarks with ones that you subsequently claimed to be outdated and problematic, anyway, because they aren't properly standardized. So, what exactly are you going with, here? This game is inadequate for benchmarking?

 

[Bacon1]

55 degree FOV is not standard, right? It looks like AMD cherry picked the exact situation where they could maintain ~60 FPS minimum on ultra, which is pointing to a card below Titan XP, assuming drivers aren't able to close the gap.

55 is default yes.

 

[Jessy]

You challenged those benchmarks with ones that you subsequently claimed to be outdated and problematic, anyway, because they aren't properly standardized. So, what exactly are you going with, here? This game is inadequate for benchmarking?

It's confusing because it is the vertical FOV (and not horizontal like many games).

It's also the default setting in this game.

 

[Headfoot]

You should probably stay more current on news. All of this has been discussed on Anandtech and several other forums and/or tech press. AMD sent Vega off to manufacturing in late June. That was the Vega party Raja tweeted about. A new Fury GPU hardware ID showed up in the drivers in late October/early November. An AMD rep responded on Reddit that the Vega demo during New Horizons was using a slightly modified Fury. Basically just enough modification to get everything to run.

I've seen nothing saying they only got first silicon back in November/December. Link or it didn't happen.

 

[HurleyBird]

55 is default yes.

Ah, I stand corrected then. Disregard my original assessment (might still be the case of course, but no damning evidence for it being so).

 

[Krteq]

55 vertical FOV is standard for all DICE games. It's 85° hotizontal FOV.

 

[happy medium]

So I dont have to read through the whole thread, how fast is the Vega looking now? Last I checked, it was 10% faster than a old gtx1080 using a game thats usually 15% faster on AMD cards. Is that where its performance still stands?

Still releasing ~ June?

 

[Bacon1]

So I dont have to read through the whole thread, how fast is the Vega looking now? Last I checked, it was 10% faster than a old gtx1080 using a game thats usually 15% faster on AMD cards. Is that where its performance still stands?

Still releasing ~ June?

Since when was battlefront 15% faster on AMD cards?

http://www.techspot.com/articles-info/1209/bench/Battlefront_01.png

http://www.techspot.com/review/1209-nvidia-geforce-gtx-1060/page6.html

We see it running VSync Ultra @ 4k which is somewhere around Titan XP range. Not sure exactly because it is capped @ 60 and direct scene comparison hasn't been done. Also running on Zen.

Not sure where June comes from since there has been no release date.

Why not just wait 12hr?

http://ve.ga/

 

[HurleyBird]

He was talking about Doom.

 

[happy medium]

when was battlefront 15% faster

It was some other game, I think it was Doom?

Why not just wait 12hr?

Oh wow ok . so they are gonna spill the beans about Vega?
thanks.

Funny Nvidia spills the beans about gtx1080ti tonight and AMD does Vega tomorrow. that's great news

 

[guskline]

It was some other game, I think it was Doom?



Oh wow ok . so they are gonna spill the beans about Vega?
thanks.

Funny Nvidia spills the beans about gtx1080ti tonight and AMD does Vega tomorrow. that's great news

Well the Nvidia livestream just ended and the Nvidia CEO did NOT announce the GTX1080TI.