• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Mantle: Teething Problems (Anandtech R9 285 review showing the downsides of low level APIs)

I'm sure this isn't what was being said a while ago.
#differentgravey and all that.

Wasn't the point of Mantle to be best. If we just wanted another option we had OpenGL. That's not in many games and doesn't perform as well as DX either :)

Its not an issue, we have choices, its an optional extra, use that option when its better.

Simple :)
 
I think this Anandtech article is total and utter garbage.

It is also a very old story and very misleading.

As most people know I have probably been Mantles biggest critic on these forums but not today.

The reason this is a very old story is it is well known that Mantle does not perform as well in some circumstances on an IB-E 6 core CPU where VRAM is limited. It has got nothing to do with the fact that they tested on a Tonga based GPU, this problem also applies to Hawaii GPUs. If people remember the number of posts I have made in the past about garbage performance @4K using Mantle on a 4930k with my Hawaii GPUs, it is the same problem. The only difference is Anandtech are using a card with only 2gb of VRAM at a lower resolution.

The question also is why did Anandtech choose to use a 4960X to demonstrate this problem, Mantle is designed and aimed at lower speced CPUs.

Until Anandtech do some proper tests with a wide variety of lower spec CPUs, this article is next to useless. An IB-E CPU will run DX11 well through sheer brute force.

It would also be interesting to see what would happen with 2,3 and 4 way CF on Tonga cards as Mantle tends to give higher fps than DX11 in these configurations as even an IB-E CPU will hit bottleneck limits quicker on DX11. Unfortunately the best that could be tested is 2 way CF at the moment as the R9 285 does not run anything more IIRC.

As I said earlier I am not a big fan of Mantle as I use 4K but for lower resolutions and low/mid range CPUs it does a fantastic job and deserves a lot of credit for this. The bottom line though is -

If you find that Mantle does not work well in the circumstances you are using to game, you will still get great performance from DX11 on your AMD card.
 
^

Problem with that being that it was AMD themselves who said that these issues are about the games being launched before Tonga and not being optimized for it.

If it was memory issues AMD could have just said something along the lines of "we'll tighten up the memory management of mantle and everything will be fine after a while".

But they didn't say that, they said this:

In explaining the situation, AMD tells us that this is an application level issue due to these games not being familiar with Tonga, and that this can be fixed through further patches.

They specifically say that fixing these issues will require extra patches from the game developers. No mention of memory.
 
They also said a long time ago that support for games in Mantle is down to the Devs, this is a very old story.

It is hardly surprising that NVidia don't want anything to do with Mantle, if a slight tweak in architecture can leave the game Devs wrong footed/can't be bothered. If NVidia with an entirely different architecture were to use Mantle then things would be horrendous with the game Devs not keeping up with new card releases.
 
So is the argument now that if Mantle doesn't work we go back to DX?
Wasn't DX getting slated as being evil and holding back performance and being the worst thing ever as Microsoft only care about consoles? And now all of a sudden Mantle is OK because you can still use DX instead?

That seems like a rubbish way to push and API, telling everyone they can just use the competing API so it's all fine!
How Mantle perform- good or bad has nothing to do with the fact that the "current dx" is lacking, and people and developers alike been crying out for something better or improvement. People were stuck with dx even if they weren't happy with the limitation, because they didn't have an alternative choice option.

Just think of dx and Mantle being the motorway and the toll...I'd rather have the option of toll available than not. If there's maintainence work on the toll leading to longer travel time than the motorway, I'd just use the motorway instead. It is just that plain and simple.
 
Last edited:
This is why it annoys me when people throw stones at Microsoft. Giving the developer more control doesn't necessarily mean gains across the board. There is still a lot of resource intensity from AMD required. Maybe if they put as much effort into it as some of their PR spin it could almost be potentially viable.
 
Last edited:
This is why it annoys me when people throw stones at Microsoft. Giving the developer more control doesn't necessarily mean gains across the board. There is still a lot of resource intensity from AMD required. Maybe if they put as much effort into it as some of their PR spin it could almost be potentially viable.

The game Devs practice the Ferengi Rules of Acquisition, rules no.1 and 3 springs to mind here.:D

1. Once you have their money, you never give it back.

3. Never spend more for an acquisition than you have to.

Or putting it another way once a game has been launched and most people who are going to buy it have done so in the first few months, the game Devs are not going to waste money on an API updating it for new hardware once they have got your money.

http://en.memory-alpha.org/wiki/Rules_of_Acquisition

:D
 
:D

It's not just the developers fault though, if at all. If someone offered you several million to help propel your game you'd take it! Just look at GameWorks, which is as it stands thanks to Ubisoft as equally as flailing if not worse in some respects.

I think the ball falls firmly in AMDs court with this one. Where there are certainly some great people there, their driver team are an absolute nightmare. I witnessed this first hand not so long ago and I'm pretty sure the issue is still unresolved. Ask Unwinder!

He might be miserable, but he'll certainly tell you what he thinks. On the flip side though progress is progress and there has been a lot of it in the last 24 months, I'm just not sure Mantle is in anyway part of that personally.
 
Last edited:
I'm curious, hasn't MS been thrown so much mud at them for DX, yet Mantle failures are dev's fault ?

Convenient:
fitting in well with a person's needs, activities, and plans
 
Like Kap pointed out on other page. We would need to see how the games perform on a lower 4 core CPU's
Maybe mantle isn't ready for Intel 6 core?

We need more testing on this before we can, call facts.

Go back and read kaps post I think he nails it.
 
There's no way developers are going to release a game patch off their own bat every time AMD release a new chipset so this another nail in Mantle's coffin.

The main reason why so many developers have snubbed PC for consoles in recent years (aside from console games being more expensive/profitable) is to avoid having to do all of the technical support for so many hardware configurations. Mantle support would make that even worse as it stands.

AMD need to wake up to the real world and accept that it's THEIR OWN responsiblity to optimise games for their products, releasing an API in the hope that developers are going to do it all for them is wishful thinking.
 
Like Kap pointed out on other page. We would need to see how the games perform on a lower 4 core CPU's
Maybe mantle isn't ready for Intel 6 core?

We need more testing on this before we can, call facts.

Go back and read kaps post I think he nails it.

Shifting gears, for the launch of the R9 285 AMD is advising reviewers and users alike that Mantle performance on Thief and Battlefield 4 is not going to be up to snuff right now.
So AMD saying it isnt enough of a fact?
 
Back
Top Bottom