• 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.

The AMD Mantle Thread

Page 117 - 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.
Anyone who's interested they gave away a bunch of stuff on Kimmel. GPU's, consoles, gift cards, etc...

I figured it was going to be something more than just a PR stunt.
 
Anyone who's interested they gave away a bunch of stuff on Kimmel. GPU's, consoles, gift cards, etc...

I figured it was going to be something more than just a PR stunt.

To the audience?

That kind of sucks if it was. I mean I bet a big majority of the people there probably don't even play video games. What a waste.
 
Context is:
you link to a person first post at b3d. How trust worthy is that for a starter. Excactly when nv pr machine is in full gear. 😉
Next time better tell that and link to the entire thread.

But you are welcome to explain to me how you interprete the results and the premises and asumptions for them. Will you care to do that?

Secondly all the graphs is just worthless because nv cards simply can not do it in actual games. And never had. And never will. Because of dx that happens to be what is used, and mlst of all because of real world programming limitations. Fact.

Cgn will be 30 times as fast for drawcalls. But only with mantle. What this thread is about.

Wow, you are slow person don't ya? Everything is showed about Matle come from AMD PR department.

Infraction issued for personal attack. One month ban added for repeated forum violations.
-- stahlhart
 
Last edited by a moderator:
Context is:
you link to a person first post at b3d. How trust worthy is that for a starter. Excactly when nv pr machine is in full gear. 😉
Next time better tell that and link to the entire thread.

But you are welcome to explain to me how you interprete the results and the premises and asumptions for them. Will you care to do that?

Secondly all the graphs is just worthless because nv cards simply can not do it in actual games. And never had. And never will. Because of dx that happens to be what is used, and mlst of all because of real world programming limitations. Fact.

Cgn will be 30 times as fast for drawcalls. But only with mantle. What this thread is about.

Please stop calling it "cgn" it's "gcn"
 


Intresting part ~37+ minuets into it.

attachment.php





StarSwarm engine with mantle and with directX.

DirectX bottlenecks ~15k batches with best CPU on market, performance plumets by huge amounts after that point.

on mantle, you can do 80k-100k without issue.



"At application level, starswarm is 3 faster in some cases"


Imagine running directX and getting 20 fps in a game, compaired to running mantle and getting 60 fps.

A 3 times speed up is huge, certainly more than just the ~20% some people have been saying it ll likely only amount to.
 

Much more informative and understandable when the slides are contextual with the presentation.

Three things jumped out for me.

1. That for all the performance improvements already highlighted, Oxide categorically states it has just 'scratched the surface' of what Mantle will enable.

2. That Mantle is enabling game development to be massively easier and more streamlined. It just flat out eliminated massively time consuming programming necessary to achieve a fraction of what can easily be realized with Mantle.

3. The last slide, titled 'Point of No Return' appears to me an implicit statement Oxide has little to no interest in implementing Microsoft's API and massive interest in taking advantage of the possibilities opened by Mantle.

DICE stated in a slide future development and optimizations for Frostbite would be focused on the PS4 and Mantle.

Mantle will be so tremendously straightforward and open ended compared to working with other APIs it's going to be mental drudgery torture for the game teams to drive Microsoft's Yugo API after experiencing the Formula 1 thrill ride of Mantle.
 
Last edited:
What does the rumor mill say about the Mantle patch for Battlefield 4 not being out yet? Did they run into some last minute bugs, or are the performance gains not as good as promised?
 
Status
Not open for further replies.
Back
Top