Vega/Navi Rumors (Updated)

[krumme]

Yes, I can see the lineup of excuses is long. Not the right drivers, its really a dual card for 300W and so on and on

Vega performs as expected when you look at Polaris.

This is just flamebait and trolling. Shintai is not stupid and knows its not a dual card.
He is just intentional provoking people and have been in nearly all threads he participated in.

 

[Dygaza]

How can this be "big" Vega? The card runs a modified Fiji driver,so is apparently using 4096 shaders which that leak said a while ago. This looks like a modified Fiji GPU with more memory and a higher clockspeed.

It doesn't make any sense that 4096 shaders would be big Vega. There isn't really much room for all other cards between Polaris lineup and Vega lineup then. I mean if there's going to be 2 cards from both Vega 10 and 11.

 

[Trumpstyle]

Amd really needed this card to be 20% faster than 1080, right now looks to be around the same performance level. Which will force them to launch this card for around 400$. They need to make money and 400$ must give them terrible profit margins.

 

[stahlhart]

This is just flamebait and trolling. Shintai is not stupid and knows its not a dual card.
He is just intentional provoking people and have been in nearly all threads he participated in.

Your account has an option to ignore members, if their posting troubles you that much.

You don't have the right to silence the opinions of others, if those opinions are not in violation of forum posting rules. If a statement is factually incorrect, show why, with appropriate citations if needed. If no agreement on a point of contention is reached, agree to disagree and move on.

You aren't getting a one-sided echo chamber here. Deal with it.

-- stahlhart

 

[PhonakV30]

Your account has an option to ignore members, if their posting troubles you that much.

You don't have the right to silence the opinions of others, if those opinions are not in violation of forum posting rules. If a statement is factually incorrect, show why, with appropriate citations if needed. If no agreement on a point of contention is reached, agree to disagree and move on.

You aren't getting a one-sided echo chamber here. Deal with it.

-- stahlhart

Question : Why didn't you warn ShintaiDK ? It's started with his Post.




Mod callout. If you have issues with the moderation, make an MD thread.


esquared
Anandtech Forum Director

 

[happy medium]

Question : Why didn't you warn ShintaiDK ? It's started with his Post.

You don't have the right to silence the opinions of others, if those opinions are not in violation of forum posting rules

I suggest you don't not derail the thread, stahlhart wont like it.

Here is the reason.





Quit trying to play mod here.


esquared
Anandtech Forum Director

 

[guskline]

Big Vega is not in my sights but I wonder how a single smaller Vega will compare to my RX480s in CF?

 

[itsmydamnation]

I suggest you don't not derail the thread, stahlhart wont like it.

Here is the reason.

It takes a lot more effort to refute BS then it does to create BS, maybe we should all just starting writing BS

 

[happy medium]

Big Vega is not in my sights but I wonder how a single smaller Vega will compare to my RX480s in CF?

"AMD Vega 10 With 8 GB HBM2 – Device ID “687F:C1” Confirmed"

"Also, for those wondering what the secret 687F:C1 product that showed up in AOTS database a few weeks back was, it is a graphics card based on the Vega 10 GPU. Actually, this is the same card that was seen in AMD’s demo PC at the Tech Summit conference. This variant packs 8 GB of HBM2 and is more focused at desktop PCs rather than HPC market. Performance of this card was on par with GeForce GTX 1080 but drivers have not been optimized for it yet."
http://wccftech.com/amd-vega-gpu-nvidia-pascal-gp100-gpu-comparison/

I would say the small Vega would be much slower than your xfire setup based on the information above. AOTS is favored by AMD cards and still looks to be equal to gtx1080. Unless small Vega is only a bit slower than big Vega 10, your xfired 480's will be faster than small Vega with good scaling.

 

[Snarf Snarf]

Doesn't 480 CF perform around ~1080? That would mean whatever this Vega shown in Doom should be pretty close to RX480 CF set up.

 

[.vodka]

https://twitter.com/GFXChipTweeter/status/808384288530067456

Raja says MI25 is single GPU.

 

[happy medium]

Doesn't 480 CF perform around ~1080? That would mean whatever this Vega shown in Doom should be pretty close to RX480 CF set up.

that's big Vega, Guskline asked about small Vega vs 480 crossfire.

 

[itsmydamnation]

Depending how big the hardware change is to cause the GFX IP level change with vega (some of the patients that are around). There could still be quite some performance to extract before release.

 

[Gikaseixas]

This seems to be small Vega to me. Big Vega will most likely be a 300w moster but i could be totally wrong.

 

[zinfamous]

that's big Vega, Guskline asked about small Vega vs 480 crossfire.

Is it?

The MI25 was shown in those slides to be significantly outperforming Titan XP --if those are to be believed--and that is a Vega chip.

I doubt this is the same Vega in the Doom demo

 

[happy medium]

This seems to be small Vega to me. Big Vega will most likely be a 300w moster but i could be totally wrong.

if Vega 10 is big Vega, then this IS big Vega.

 

[happy medium]

Is it?

looks like to me.

"AMD Vega 10 With 8 GB HBM2 – Device ID “687F:C1” Confirmed"

"Also, for those wondering what the secret 687F:C1 product that showed up in AOTS database a few weeks back was, it is a graphics card based on the Vega 10 GPU. Actually, this is the same card that was seen in AMD’s demo PC at the Tech Summit conference. This variant packs 8 GB of HBM2 and is more focused at desktop PCs rather than HPC market. Performance of this card was on par with GeForce GTX 1080 but drivers have not been optimized for it yet."
http://wccftech.com/amd-vega-gpu-nvidia-pascal-gp100-gpu-comparison/

"The folks over at pcgameshardware.de have also confirmed that this is the very same graphics card “687F:C1” that we spotted mingling with other GTX 1080s on AOTS’s benchmark leaderboard a couple of weeks back."

http://wccftech.com/amd-vega-performance-8gb-hbm2/

 

[Arachnotronic]

if Vega 10 is big Vega, then this IS big Vega.

Vega 10 is probably big Vega. Dreams of a larger Vega are pipedream, IMO.

 

[Gikaseixas]

Vega 10 is probably big Vega. Dreams of a larger Vega are pipedream, IMO.

Yep, no one knows for sure. People thinking that it is big Vega could be pipedreamers too
Good thing that is confirmed - it is a single GPU and not dual GPU like it was being rumored

 

[krumme]

Vega 10 is probably big Vega. Dreams of a larger Vega are pipedream, IMO.

Agree. How do you think the amd consumer gpu portfolio is mid 2017 also positioned vs nv? - i simply cant find room for 2 dies above p10 both stuck in a bad place as i can tell. Makes no sense to me unless its some tsmc whatever...wsa stuff interfering?

 

[linkgoron]

Agree. How do you think the amd consumer gpu portfolio is mid 2017 also positioned vs nv? - i simply cant find room for 2 dies above p10 both stuck in a bad place as i can tell. Makes no sense to me unless its some tsmc whatever...wsa stuff interfering?

The same way Nvidia has two dies in their stack above GP106.

 

[zinfamous]

looks like to me.

"AMD Vega 10 With 8 GB HBM2 – Device ID “687F:C1” Confirmed"

"Also, for those wondering what the secret 687F:C1 product that showed up in AOTS database a few weeks back was, it is a graphics card based on the Vega 10 GPU. Actually, this is the same card that was seen in AMD’s demo PC at the Tech Summit conference. This variant packs 8 GB of HBM2 and is more focused at desktop PCs rather than HPC market. Performance of this card was on par with GeForce GTX 1080 but drivers have not been optimized for it yet."
http://wccftech.com/amd-vega-gpu-nvidia-pascal-gp100-gpu-comparison/

"The folks over at pcgameshardware.de have also confirmed that this is the very same graphics card “687F:C1” that we spotted mingling with other GTX 1080s on AOTS’s benchmark leaderboard a couple of weeks back."

http://wccftech.com/amd-vega-performance-8gb-hbm2/

you posted that very same thing 4 posts prior to this.

what does it accomplish, other than neither confirming nor rejecting your argument..for the 2nd time?

 

[HurleyBird]

It's not worst case or best case, but probably the best showing AMD can come up with right now. I don't see any proof that the intake fans are taped over, and I can't see AMD purposefully throttling their cards, so I'm rejecting that conjecture. At the same time, these chips are apparently fresh out of the fab and using very unoptimised drivers.

I can see performance improving a fair bit with proper drivers, but this also means that Vega is probably still some ways off. To be honest, they'd probably be better off having this level of performance and launching around year end then they would an additional +10-20% and launching in Q2.

As far as big versus small vega, I give it 50/50. On the one hand, having only 8GB of memory and the existence of Polaris 12 (which would slot nicely between this and Polaris 10) point towards this being small Vega. However, AMD's pattern of inept execution and shooting themselves in the foot cannot be ignored.

 

[Vesku]

If this is big Vega there is little reason for AMD to produce little Vega. Unless little Vega is GDDR5/X then there would still be a reason to sell it between P10 and big Vega. Also this Doom performance preview looks to be about where a Polaris GPU would hit with 9-10 TFlops. With the Doom 4K demo we either have a small Vega with better gaming/TFlop ratio than Polaris or a perf/W optimized big Vega demo where it's gaming/TFlop is barely changed from Polaris.

 

[happy medium]

you posted that very same thing 4 posts prior to this.

what does it accomplish, other than neither confirming nor rejecting your argument..for the 2nd time?

I was using common sense. I was letting everyone know that the same card they used to test AOTS 3 weeks ago, was the same card they used for the Doom benchmarks in todays article. I was not arguing, its a fact.

We all know big Vega to be Vega 10, correct?

Quote:
"AMD Vega 10 With 8 GB HBM2 – Device ID “687F:C1” Confirmed"

quote:
Vega’s Confirmed Specs
Members of the press inside the demo room were able to spot some key specifications pertaining to Vega by taking a look at the expanded statistics in Doom. 8GB of HBM2 memory for the consumer version of Vega was confirmed.

So in conclusion what we are seeing IS big Vega 10, correct? Yes it is.
Next conclusion we come to after seeing the 2 leaked benchmarks is that this card is Vega 10, otherwise known as big Vega and is a little faster than a gtx1080 in AMD favored games.

It seems rather simple.