• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Vega/Navi Rumors (Updated)

Page 83 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
AMD confirmed there will be 4 GB configurations
https://videocardz.com/67642/amd-radeon-rx-vega-is-just-around-the-corner
And according to demos we've seen, it should be enough due to HBCC. The only issue is marketing 🙂

I get the strong impression from the article that mainly seems to be aimed at laptops that usually have a fixed resolution when not using an external screen.
So, for a 1080p screen 4GB might be enough. At least for older games. One might never know what future games will do and if the caching method will be enough.
 
I genuinely suggest reading people which are in higher understanding of GPU architectures, than you are, because you are spawning BS, that makes you look like an "...", attached only to your view of the world. There is enough analysis done on Beyond3D forum, by people from game development. Do your research, then post.
that should be an actual sticky post over every single thread in this subforum
 
LMAO, please tell us more since you're one of those guys that has ''higher understanding of GPU architectures''...
After reading your second quote it seems like you're the one spawning Epic BS...Without a single proof....
then again, how would you know if he does not have any proof?
 
Is it possible for a company to have access to something like that which belongs to its competitor?

I'm sure it's possible. But I believe that is also corporate theft, and is a felony? I don't think anything like that is going on. I know this is line of thinking is silly in this thread, but the reality is that nVidia has absolutely no need for such shenanigans. They are perfectly fine worrying about their own hardware.
 
What is interesting is that AMD more promotes streaming in what is needed.
That got me wondering about euclideon, i wonder if the techniques vega uses can be used together with how euclideon can be used to create virtual high reality worlds that are terabytes in data size.
Euclideon is a streaming 3d engine that can use laser scanned areas that actually exist.

https://en.wikipedia.org/wiki/Euclideon

 
No, it was a sarcastic comment. If any proof of such a thing would come to light, any company would be in MASSIVE legal trouble. No one would risk it.
That would depend on:
a) if you're found out
b) if you have pockets deep enough to cover it up

Think 2007 Spygate if you follow Formula1.

But yeah I get your point.
 
I just read the Anandtech optane P4800X card review. I think that is what a video card needs. I think with that type of memory we could not only cut out the need for so much VRAM but also make games load a lot faster.
 
What is interesting is that AMD more promotes streaming in what is needed.
That got me wondering about euclideon, i wonder if the techniques vega uses can be used together with how euclideon can be used to create virtual high reality worlds that are terabytes in data size.
Euclideon is a streaming 3d engine that can use laser scanned areas that actually exist.

This seems like the perfect job for a Radeon SSG Pro card. 2TB M.2 PCIe SSD to hold all that scanned data getting rendered in real time to a VR headset. I'm getting excited thinking about it :smirkcat::smirkcat:
 
This seems like the perfect job for a Radeon SSG Pro card. 2TB M.2 PCIe SSD to hold all that scanned data getting rendered in real time to a VR headset. I'm getting excited thinking about it :smirkcat::smirkcat:

I see your point.
And imagine if the image quality would be like Mizuchi.
That would be some experience indeed. 🙂
 
That would depend on:
a) if you're found out
b) if you have pockets deep enough to cover it up

Think 2007 Spygate if you follow Formula1.

But yeah I get your point.

https://www.extremetech.com/extreme...t-former-employees-for-alleged-document-theft
http://www.crn.com/news/components-...four-former-employees-who-left-for-nvidia.htm

Never say never.
Transferring 10000 documents to nvidia is no joke. who knows if nvidia had their next gcn roadmap and knew what was coming for them.
 
That would depend on:
a) if you're found out
b) if you have pockets deep enough to cover it up

Think 2007 Spygate if you follow Formula1.

But yeah I get your point.
When I was first reading it Spygate came to my mind first. But I think the Coke recipe with Pepsi is an even bigger example. Here you have a Trade Secret, not actually registered IP, the reason being that those would expire and even if for the secured period of time people would know how to make it themselves. But even as a trade secret is so protected that when someone tried to sell Pepsi the Coke recipe, Pepsi was so worried that instead of just saying no, they called the feds and set up a sting. Both of these companies have deep ass pockets, but Pepsi wasn't going to take any chances.

Stuff that actually happens is more like Spygate. New company hires old company, doesn't realize that the new product they are developing is born from work the guy did with the previous company. It's why Newey was on gardening leave for like 3 years before he signed with a different team. Murray as well before him went on gardening leave, was allowed to have some fun and designed Mclaren F1 and decided he just wanted to create road cars after that.
 
Someone refresh my memory.. wasn't there the same lack of leaks prior to Fiji's reveal? I think most here know I want AMD to succeed and I DO believe Vega has some very promising technologies that may put AMD back in the lead but I am not taking AMD's silence as an indication of its performance, good or bad.
 
I'm sure it's possible. But I believe that is also corporate theft, and is a felony? I don't think anything like that is going on. I know this is line of thinking is silly in this thread, but the reality is that nVidia has absolutely no need for such shenanigans. They are perfectly fine worrying about their own hardware.
That's right, corporations never do illegal things.
I just love the blind trust people have of companies......
 
There is some serious larping happening in this thread. Last year before the RX4XX release a similar thing happened over at OCN with some guy claiming that polaris couldnt validate above 850. Id be wary of anyone claiming inside knowledge of vega. More than likely they're just seeking attention.
 
where did you get from my comment "corporations never do anything illegal"?

wha?

I think you've jumped the cliff and landed in the sar chasm.

From the linked article, there is some reason to believe it's possible as AMD accused 4 employees of doing essentially what is being suggested, giving confidential information to NVidia. I tried to see what came of the case, but after the initial reporting, there doesn't see to be much of any fresh information.
 
All this back and forth about PCIe bandwidth not being sufficient for the HBCC and memory paging system takes the word streaming way too literally. The design does not imply that textures and frame buffer are CONSTANTLY streaming from PCIe to system RAM. The data from the frame buffer is intelligently managed in real time, and if new assets for future frames need to be called, they get called BEFORE the frames need to be rendered. PCIe 16x 3.0 bandwidth is more than sufficient to handle tiny calls for data.
and AMD invented time machine to predict the future and what the user will do in next frames... seriously ?
this topic becomes ridiculous 🙁
 
and AMD invented time machine to predict the future and what the user will do in next frames... seriously ?
this topic becomes ridiculous 🙁

CPUs have been doing branch prediction and cache predictions for decades.

There's no particular reason a GPU must load every texture into VRAM just in case. In fact from a statistical stand point most frames look almost the same from refresh to refresh.

Based on AMDs memory work my gut feel is that even with a smaller amount of VRAM Vega will not choke at any resolution and settings a Titan Xpp doesn't.
 
Status
Not open for further replies.
Back
Top