AMD and Electronic Arts send word this afternoon that the Mantle update for Battlefield 4 has been delayed until next month. The update was previously scheduled for late December, however any slippage on that schedule would push the release to January, which it appears is exactly what has happened. AMD and EA have released a short and highly sanitized statement on the matter.

After much consideration, the decision was made to delay the Mantle patch for Battlefield 4. AMD continues to support DICE on the public introduction of Mantle, and we are tremendously excited about the coming release for Battlefield 4! We are now targeting a January release and will have more information to share in the New Year.

With that in mind, for Battlefield 4 players it’s well known that DICE is in the middle of a massive bug hunt due to a number of recurring (and sometimes severe) bugs in the game, which has led to DICE pausing most other development tasks in order to focus on fixing bugs. As such there’s been a lot of speculation over whether Mantle would be delayed as part of the bug hunt, and to minimal surprise this seems to be what has happened.

With that said, while Electronic Arts’ statement is unfortunately (but not unexpectedly) light on details, given the compartmentalized development of modern engines and the bugs facing Battlefield 4 we have good reason to believe that Mantle development itself has only been minimally impeded (if affected at all) since the bulk of BF4’s issues are not in the rendering engine. Instead it’s far more likely that DICE and EA’s QA teams are tied up finding bugs and testing fixes, which would require delaying the Mantle update due to a lack of resources to validate it. The silver lining on all of this being that if our assumption is right, it would at least mean the Frostbite rendering team would have more time to spend on the project while waiting for QA resources to be freed up.

As for the state of the Mantle API itself, we don’t expect that this will change anything. AMD is already working with other developers on Mantle and we’ve already seen Mantle on display at the 2013 AMD Developer Summit, so we know it’s up and running in development form. But as Battlefield 4 is still going to be AMD’s launch vehicle for Mantle, this means that AMD’s consumer Mantle plans are essentially delayed in lockstep with Battlefield 4.

Comments Locked

66 Comments

View All Comments

  • TheJian - Tuesday, December 31, 2013 - link

    Considering you were totally wrong on consoles using Mantle at all, YMMV.

    At most according to one dev you'll get 20% from this which can be had with a simple gpu refresh, driver updates etc. There is no point in a dev spending on this without AMD refunding whatever they spend. Mantle gets a dev ZERO extra dollars from a game. You are better off making a BETTER game with that time that will get higher ratings and thus MORE SALES. Mantle gets you a few extra happy AMD users (by few I mean only a portion of AMD's cards can do it and most are going to miners paying dumb pricing).

    A simple driver update can net half of what this does easily. We see 10%+ on many driver updates across many games per update. Also note AMD said "we wouldn't bother for 5%". With a dev saying 20% isn't "unreasonable" I don't expect 20% very often. With AMD not saying "we wouldn't bother for 10%" you can expect 10% a lot with no dev claiming more than 20% and ONLY ONE claiming even that as a MAYBE case. It isn't exactly a glowing comment saying "not unreasonable". If it was to be expected to be the norm you'd say something more like "expect this a lot". I understand why AMD likes a proprietary tech (It's proprietary until its is RUNNING on something other than AMD no matter what they claim - NV will never make a GCN core and we have no proof it runs on anything else), but have no idea why a dev would waste time on it given it make no extra cash from doing it. IF they could charge an extra $10 to AMD users, ok maybe it was worth the dev time/resources spent. But that isn't the case. So why bother with what can be had with a driver update or two, and at worst the next refresh of vid cards? In one rev of cards you get EVERYONE more perf and ask devs to do nothing to get it. This Mantle way, every time they have to squash bugs etc Mantle will go back burner. Instead of Mantle, AMD should have been working on a Gsync tech that requires nothing from devs. I think the same of trueaudio. What a waste.

    I can't even believe you thought Mantle would be in consoles. MS will not allow competition for DirectX on purpose...ROFL. Too much AMD fantasy and speculation on this site these days. Your mantle article was 3 parts fantasy fanboy dreams and 1 part reality.

    Remember people, engine support means nothing. A dev still has to USE that part of the engine. You don't have to use physx just because unreal engine 3 supports it and with consoles blocking Mantle, you have less of a chance of success than physx which IS in consoles. Top that with Mantle's launch vehicles being in the hands of the worst company on the planet for the last 2 years (maybe a 3rd win this year with a BF4 lawsuit now from shareholders, surely BF4 owners will follow soon in their own suit), and you have a recipe for disaster that we see now. Then add to that EA foisting Frostbite on all their games (even where it doesn't belong), and well, you should expect some bad games coming soon.

    Clearly EA is planning on maximizing the Mantle optimizing or they're just getting even dumber than the last 2 years. A one size fits all engine is not a recipe for success or happy customers. There's a reason you make say, the Infinity Engine for an RPG but don't use it for your racing games. EA can't go bankrupt soon enough for me (and replaced by 100 mini-devs who make GREAT GAME PLAY games, instead of 10hr or less PRETTY games). Between EA, MS and a few others they destroyed most of my LOVED companies of the 80's and forward (Origin, Bullfrog, Interplay etc etc the list is long that these people have slaughtered). Gobble up awesome company X, force out a turd before it's done (like BF4) for users to beta test for a year, then patch it to death for another year with updates and DLC culminating in a ULTIMATE or GOTY edition that should have been the ORIGINAL release for $40 or less and lasts twice as long or more. I get no printed manual, no disc, no returns and $60 for the main game + $20-40 for DLC and even more if microtransactions are included. Just say no to $60 games you can finish on Saturday. I remember when they said no manual/disc/box would SAVE me money. Of course I didn't believe it.

    Also, I think this will wind up being one of AMD's biggest mistakes. Creating Mantle without the market clout to push it, or the money to pay to get it used (30% vs. 65% for nv in market share), merely caused a response from NV that actually CAN be funded with 2.7B in the bank and no debt.

    I doubt NV would have made Gameworks if AMD hadn't pushed their own proprietary game tech. NV can spend on 10 engines for every engine AMD can afford and still make money. AMD on the other hand has to make 200mil+ just to cover interest on their own huge DEBT before any investing in future tech can happen (or you just run red for life I guess). They have owed fines to GF yearly for years. This years is 200mil due TODAY Dec 31. Get ready for another AMD "one time every year charge". I'm confused by a charge that is ONE time happening YEARLY...ROFL. With consoles chips NOT being made at GF (rather TSMC) there is no chance AMD will avoid another fine for next year. Take or pay sucks and you should be fired for signing up for this crap. Verizon/AMD should be examples of what NOT to do in Business 101 class next year :) This same type of agreement may cost Verizon 14Billion due to lacking iphone sales. Apple can't just let this ride, or everyone else will say "but wait a minute"...Even if they let 1/2 of that slide, how many other companies (which are coming up short on iphone sales all over the globe) will be asking for half of their contracts to be forgiven? Even if Apple (or GF for AMD) wanted to let stuff slide they really can't risk the domino effect that will come seconds later.
  • siliconwars - Tuesday, December 31, 2013 - link

    I've read a lot of shit in these comments in my life, but this is on another level. 10%? Lol you have...oh dear you have absolutely zero clue.
  • HisDivineOrder - Tuesday, December 31, 2013 - link

    You realize that draw calls aren't always the limiting factor with regards to graphics performance, right?

    From what I read, most developers did think 20% was optimistically the high end and that lower performance improvements were considerably more likely. Then you have Tim Sweeney and Carmack, both coding greats in their own right, that look at this technology and seem completely against it.

    Carmack may have retired, but Tim Sweeney's still coding the most pervasive gaming engine out there for the games we care about and it's important what he thinks.

    Odd how most of the developers who say Mantle is going to net them that any gain in performance (20% or less) are all ones with some kind of partnership with AMD. It's almost like AMD promised them money and/or technical support to use Mantle. As if Mantle by itself is not enough incentive by itself.

    I'd say the same of PhysX, btw. It's not compelling enough without nVidia's promise of extra technical support to get it in there.

    Except PhysX defaults to CPU mode for non-nVidia GPU's, which means the API actually works for people with AMD cards or Intel integrated GPU's. Mantle's all AMD right now and it doesn't look like that's going to change anytime soon, either. Nor has AMD made a formal attempt to make it an open standard that they don't fully control.

    So yeah. Sorry.
  • SlyNine - Tuesday, December 31, 2013 - link

    No he's right, you don't have a clue.
  • TheJian - Wednesday, January 1, 2014 - link

    It was at AMD's APU show not long ago that a SINGLE dev said it wasn't unreasonable to get 20%. Nobody else has said a single thing from ANY developer. AMD themselves say they wouldn't do Mantle for 5%...IF it was a lot higher that you were expecting to get why not shoot higher than 5%? Why not say "we wouldn't bother with Mantle for 15%"? Ahhh, right, because most of the time you'll be between 5-20%?

    Links to developers saying ANY number over 20% please. I can't find ANY from a dev. Only the guy on stage at AMD APU 2013 recently. Nobody else has said anything, and there are no benchmarks etc. It would seem hisdivineorder knows exactly what I know, which is the only thing they've said...LOL. Which pretty much sounds like, if you pray to god, do some voodoo, sacrifice a chicken etc, well...It's not unreasonable to get lucky and have 20% gains. WOW. Don't expect 20% much. One driver update almost makes that pointless. Never mind yearly refreshes of vid cards that allow everyone the 20% (far more?) with no dev doing a think except sitting back and waiting for the next driver update or two, or a card refresh. Why code more?

    If he has no clue, at the very least you have no proof he has no clue as you've provided none to refute his statements or mine. So with no proof to the contrary, I guess he still has the same clue he had before you posted ;) It appears he watched both shows like I did. He is entirely correct about carmack and sweeney (clearly didn't like it verbally on stage, even Andersson (dice) had a hard time defending it), they are unimpressed with Mantle and hope NV doesn't respond in kind. But I think they already are starting to with GameWorks libraries etc. AMD shouldn't have woken the sleeping giant until they could afford to. :(
  • yannigr - Wednesday, January 1, 2014 - link

    Performance with Mantle could vary much more than that 5-20%. It could be from that 5% when only the cpu was getting a boost, in a way, by having much less to do, and that 5% could be also on a system with Nvidia gpu, up to much higher than 20% if the complexity of what you where seeing on screen was huge, much more than what you see in PC games today. Think it like PhysX. With PhysX, maybe we see more visual fx than what we would see from any physics engine made to run on cpu. The same could happen with Mantle. Choosing "Ultra settings" in a future game that supports both Mantle and DirectX could be something much more different than what we see today going from High to Ultra. It could mean so much more complex and detailed graphics on screen that will be impossible to run the game with DirectX and anything less than an 8-12 thread Intel processor.
  • Th-z - Thursday, January 2, 2014 - link

    "Links to developers saying ANY number over 20% please."

    Second last slide:
    http://www.hardware.fr/news/13450/apu13-oxide-fait...

    "they are unimpressed with Mantle"

    Johan Andersson has repeatedly said publicly that he is exciting about Mantle. Do you know him personally and he told you otherwise in private? As for Sweeney in that NVIDIA-held event, he was curious about Mantle and acknowledged that there are a lot of overheads in DX. He didn't say a low level API was not needed, just didn't want all the hardware vendors to have their own API. Carmack was less enthusiastic about it but he did say "there would be days where it would be extremely tempting", and there is difference between "unimpressed" and "not doing it right now" given he is busy with Oculus Rift.

    So I am curious where do you (and HisDivineOrder) get that they are unimpressed?

    "they already are starting to with GameWorks libraries"

    So you don't seem to have problem with NVIDIA's close implementation that wants developers to use to make their own hardware performs better. Even if it's only 5%~20% gain? So why do you have problem with AMD wants developers to use Mantle? Is this really about fanboyism?
  • hero4hire - Friday, January 3, 2014 - link

    It's like arguing with my ex. Some facts sprinkled with crazy and a ton of conjecture in an overwhelming negative opinion. To me it's an api for devs to take or leave. If they take, with or without AMD funding, and I get 1% increase, then yay for me. I don't care about either companies bottom line, business plan, or debt rating when I buy a video card. It's for fun! Not a 10year prospectus.

    If AMD goes out of business and all there employees never work again after 2015, will that crush me and my $200 fun time hobby that I bought to see flashing triangles dance about?
  • bwat47 - Saturday, January 11, 2014 - link

    All of these numbers you are quoting are completely wild speculation. When we have some game using mantle, there will be proper benchmarks and comparison. Until then, we just have moronic flame-wars between nvidia and amd fanboys both spouting nonsense numbers out of their backsides.
  • jardows2 - Tuesday, December 31, 2013 - link

    Correct me if I'm wrong, but didn't the game developers ask for someone to do something like Mantle? And AMD was the only company interested?

Log in

Don't have an account? Sign up now