In the AMD Mantle presentation at APU13 this was specifically addressed. Mantle has different levels of abstractions and was specifically designed to accommodate future architectures.
I think I saw a slide that say, some stuff around memory is not abstracted at all, great.
Yes. I'd like to add to that, that with DX the driver is optimized to the game, or in reality, games. Dozens of them (100's?). I've heard many times people question and complain about the size of the driver D/L. With Mantle the driver is transparent so the devs can see exactly what's going on and the games are optimized to the driver. People need to think about just that one aspect for a minute and how much more sense that actually makes.
Exactly yeez!
One thing is just plainly trying to understand what they claim
Another thing is GRASPING what that (and other stuff) means in practise and what consequences it will have, what other areas will it affect, what other issues could become obsolete.
Mantle is just a big win for developer and consumer, and performance is just, let's say, one third of the benefits.
Nvidia's Denver Offload CPU is, if rumor is true, can only be just a win for consumer and even that win is debatable because by the time in roadmap 2015, assuming Q4, Mantle's going to be basically established by the time, not saying mature or widespread, but a dozen of top developers should be on it and making games more complex than ever before, will the Denver CPU then be able to compete with the added stress that Mantle cause. And totally not saying it's a bad thing, it's good, it may make old DX games work better, that could be kind of a market for retro gamers still wanting to play some of the games from the previous decade.
I'm not sure that makes any sense at all. What happens when Mantle drivers get updated, it can throw off all the games designed for a particular version. The drivers would still have to be optimized for each game when Mantle is updated. Then you are still left with what we have now, games designed while using a particular driver, and as new drivers get released, AMD attempts to improve things.
As new GPU's are released, they have to be retroactively made to work with older games.
The only new part is the lower level access.
Naaah
The old code will be there for compatability, they can simply update the game if they want to use newer features via extensions. That's one way, but if existing stuff get's updated just needs a new version. However it definitely doesn't mean the developer will have to start from scratch to make it support the new standard.
And Mantle updates most likely aren't going be that frequent like driver updates. The bulk of the support, updates, improvements would be in the validation layer and SDK, those would be more frequent, shouldn't make impact like you think.
No. Just like now, with DX compliance, the cards would need to be Mantle compliant.
And that too yes.
If that's the case then Mantle will be still born. So far it's looking like Mantle is being adopted at a much faster rate.
Rebellion was a big surprise.^_^
You aren't looking at this correctly. nVidia will need to support Mantle's feature set, just like they now need to support DX and OpenGL feature sets.
What features are missing from nVidia GPU's? Is this known?
I think they meant minimal feature set, then what each architecture has their own bonuses goes into optional exensions.
There will still be changes to drivers, which may effect performance, after a game is released. They just are more in control, and likely won't require as much assistance in the driver department.
Nope, I don't think that'll be the case at all, that's why they call it a thin-driver. They won't be doing any app-specific code in mantle driver.
Not less assistance, more like no assistance. It's a big deal something Denver CPU cannot fix, if that's what nvidia is going to do as rumored.
Nvidia users would in that case still be plagued by beta drivers for every game, having to wait for nvidia to fix them if something's wrong, etc. Mantle users will cry out of joy on first day (unless it's a rushed game like bf4)
... Mantle is approaching an order of magnitude better, and that's in the early alpha stages.
Indeed. The PC is like 10 times better than consoles in hardware (top ones, what about 4xGPU setups, eATX)
I just like that everyone doesn't realize this and keeps talking about "10 to 20%", so they'll be plesantly shocked later; that's just the very beginning. But I'm not saying that means it'll be 10 times more FPS, these numbers don't transfer linearly.
Is your claim that there is no difference in the limitations between a singlethreaded and a multithreaded DX 11 driver?
Simple yes/no.
I was thinking that as well, could be one of those "deffered contexts" things.
I wish I had more background on that, I don't, I can't say anything.
Ehh. I was looking for the "immediacy" factor. Like, when I saw crysis for the first time in 2007 it was jaw dropping. Far better than anything that proceeded it. You know? When I saw crysis 3 I was pretty darn impressed. I'm thinking to myself, if Mantle truly does have all this power at its disposal as AMD's marketing indicates, go all out. Give us something completely ridiculous in terms of pushing the limits.
A little bit offtopic ... When I first saw Crysis in some local magazine, I was much less experienced and still mostly about a 16 year old kid, so I was saying BS like "crysis sucks" for no reason at all, I think i heard some internet consensus without research at all, just others saying stuff about being "demanding", it was a prime example of group-think and low attention span, being easily manipulated. When I actually played the game, after talking how "it sucks" while I never played it, it ended up being the biggest favourite on PC for me. (I like adventure exploration, spec ops sneaky gameplay, similar to metroid and zelda)
You can probably guess I wasn't a fan of Crysis 2 at all. Crysis Wars multiplayer had so much potential but Crytek didn't give a ... leg about it.
As for the rest of your post (not quoted) the performance aspect is the thing that is most clear, so, if there are 300FPS possible in crysis 3, expect they'll use 200FPS for better visuals, depends if they target 60fps or 120.
Remember, for Oculus Rift, if it's not 120FPS it's not real VR. So 60FPS doesn't work for any serious VR.
... This whole dx11 multi-threading point is a complete nonsense.
That's what it seemd to me as well.
You seem to copy every single PR statement as the truth and hype it to no end.
Now that's a real conspiracy
theory.
You think we wouldn't have detected that by now?
Maybe you should check their history, then answer me why they wouldnt.
Marketing is marketing, they are gonna try spin the most positive cases as possible. Even going above and beyond what it can actually do.
Ohhh indeed, we do, we do know the
history.
Maybe you should check their history, then answer me why they wouldnt.
Marketing is marketing, they are gonna try spin the most positive cases as possible. Even going above and beyond what it can actually do.
And just look at Mantle so far.
Its in consoles! Ups, its not in consoles.
Its open source! Ups, its not open source.
Its super quick and cheap to implement! Ups, its not so easy or cheap after all.
nVidia/Intel can support mantle! Ups, they cant support mantle.
Conflicting statements.
BWAHAHA That is all mainstream media and casual follower misinformation, speculation, lack of attention span, and all the other reasons ...
I'm covering this for weeks now, look at
this post (cause im not going to re-type all that)
The lower level, the higher cost and extra debug time usually.
Totally untrue.
Ref:
http://www.youtube.com/watch?v=sSY2KXBoro0?t=28m14s
(from 28m to the end, they explain how big trainwreck DX development is)
Who knows? Clearly BF4 is a mess but that's not the fault of Mantle. The problem DICE has just now is that if they release the Mantle patch they'll just get accused of putting the DX game aside. It's a no-win situation.
Plus, mantle won't fix any gameplay bugs.
When is "soon"? Next month, 3 months, 6 months, more? Whats the official statement on it?
Last DICE statement was they threw in the towel and cancelled everything else but bugfixes for a good long term.
That has noting to do with mantle. It's not a DX thing, it's not a gamplay side thing, it's not a data loss thing, it's not a netcode thing, it's not a hit-box thing, it's not a balance thing.
Do you have inside information or can you substantiate this statement with actual proof, considering the state of the BF4 game being buggy? EA is presumably going to take care of the game bugs first as a priority before working on DLC or Mantle. Citation please?
If there's a reason it's a non-technical reason like mentioned above, there could have been programmers required on some deep debugging on one of the bugs there, so they could temporairly be
But, I don't think so, Dice has their support and balance teams, frostbite another 60 guys working purely on the engine and half of those are assigned to focus on some , there's a lot of layers to get to the mantle guys, the core is probably still there.
Do BF4 bugs really need graphics programmers to do the debugging in order to fix them? those bugs are mostly on the engine side, and let's keep mantle separate, when I say engine, I don't mean minus the graphics backend, I mean all the rest.
That is generally the situation for any product that has not been released yet, yes.
There is also a profession so called, researching for circumstantial evidence.
It tells me it's not a PR spin.
It tells me it's a revolution on the PC.
What I'm talking about all the responses, it's what the data is showing right now, not the PR data, but the context that goes along with it, and for that you need prior experience and knowledge of what low-level actually is and what it means in practise.
And from the post mentioned above, I've addressed the mass misinformation out there, it's the minority who understands how of a big deal Mantle is, majority are lazy consumers who have weak problem solving skills, group think (herd behavior), peer pressure, and other congnitive and mental disadvantages, which results in worthless posts like "
i'll believe it when i see it" and all the other kinds of excuses masked as "being cautious"
As I said, there is nothing wrong with being a skeptic, but this is not a banking conspiracy. If there was PR spin, I wouldn't have detected that by now.
Unless it's a giant hoax and they'll delay it for 3 months and come out laughing on 1st april, but that's not the point, the point is we're explaining based on all the data plus analysis, if it's a hoax in the end, it's not my fault. Do they have any motive to do that. Not saying it's not impossible, but AMD doesn't seem to me like it can be compared with bigpharma, big agro, crony banksters, etc
This whole PR accusation thing can also be a way to distract us... but it should I go deeper and end up in offtopic ?
I wouldn't call the oxide demo a PR show seeing that 90% of it was jargon only developers would understand.
Exactly, that's one of the points as well.
hey guys, you know that common misconception about how when you use crossfire, the vram of the cards adds up, so that two 3 gig cards would have 6 gigs total?
With crossfire under mantle, this could actually be the case, right?
Actually I heard that would be the case with mantle, but I can't confirm I heard it correctly, I would have to re-check that, but I don't remember where i saw it.
If thats the case, since the Frostbite spinoff, Mantle for BF4 is not even certain anymore.
1. Johan Andersson from DICE/Frostbite had the idea to create mantle.
2. Asked Nvidia and Intel.
2. Both denied
3. AMD Accepted
4. BF4 first test game.
5.
History in the making.
Unless nVidia adopts Mantle then I predict this will be about as successful as 3D vision or hardware PhysX. Any advantages it might have won't mean much if developers don't target it, or if it splinters PC gaming.
Right now it's looking like another Glide. DirectX may not be perfect but it's responsible or getting us out of the 1990s where every vendor from 3dfx to S3 to Rendition was pushing their own "pedal-to-the-metal, ours-is-better-and-more-efficient" API.
Yawn ... I've probably grinded through all of these many times ..
Nobody wants a different video card for each game they own.
That's consumer's problem, they want compatability, that's what they get first, not performance. They expect to get both.
I think the best way forward is for Microsoft to ship something like a DirectX "lite" which lets developers forgo some of the usual niceties in exchange for lower level access.
Why should we care about microsoft at all, do they provide anything that improves your life now that the we broke off their chain the PC industry , no, they lobby to take your freedom of speech on the internet away among other things, worthless company I don't care about let alone trying to discuss what options they have.
Nothing to you personally, but, I've addressed this elsewhere, with any tech discussion there's always these people with these thoughts at the end, and your posts just fits a classic example of this, there's is this culture almost a kind of celebrity worship you have in entertainment, but this is more of a "balance sheet worship" or "stock price worship" or "NDP numbers worship", neogaf being king of these; of avreage-gamer-debaters who debate in this semi-analyst financial manner with emotions (oh nooo my fav company lost 10 bucks i'll go cry now) about all the gaming and tech in a kind of pro-corporate pro-capitalism fashion, like we even care about their profits that is our money being milked to death. If you ask me honestly I hope microsoft just burns to the ground and Bill gates gets thrown into jail for funding poisonous vaccines that kills many children in africa. So I absolutely am celebrating to see DX being oblitarated from this planet once and for all, we don't need their system, we can build a better one that benefits everyone, not some freaks in wall street.
(Only when i've written this i've noticed your forum status, but I'll stand man of the word and keep all this intact, to clear something out, I meant to say that your post reminded me of those, not directly, also please note i sometimes use the "you" in a general context meaning "if anyone" so not you or anyone here, also just my opinion nothing more, but I don't want to make a big deal out of this so let's move back to mantle)
Some amazing predictions without even a single Mantle game being released yet.
Data analysis shows that. Not my opinion.
I would hate the days of glide and different patches depending on what graphics card you had.
That's a sacrifice I'm willing to make ANY day for top performance. I am a very different type of consumer
DX saved us from API hell...people wanting to go back to times before that, really need to take a look at history again...
How things evolve is not fixed, if they had someone come up with an idea like Mantle, DX wouldn't exist like it did, or not at all.
These points are just the kind of I'd imagine casual iPhone users would make, want easy easy eays, want simple simple simple, seamless, intuitive, bu bu be bu, all these buzzwords usually indicate something has to be dumbed down or performance sacrificed.
If you can be bothered, buy a console. No more API/OS/diver/patch for you. Not even finding playable settings. Just sit and play.
Everyone here seems to want mantle to be all vendor open standard. When it comes to nv exclusive features - some count those as added value, something that amd lacks. You have to make up your mind, you are against vendor locked features and after open standards, or you want disabling features from competitor's use. Can't have both.
Exactly, why should PC be limited by standards made up by, literally, lazy couch attitude.
They really do speak for them selfs only, I don't have a problem with many different patches, or having to do some preparation or extra step before I play, that's such a massive non-issue for the benefits elsewhere it brings, so it's a set of tradeoffs and I basicaly sacrifice a little compatability and other stuff for performance.
They see it as a disadvantage compared to before, in one of my posts, no, Mantle disadvantages are to us
ACCEPTABLE SACRIFICE, I don't consider them as downsides.
Mantle is going to go the way of Physx and end up being used in the lamest way. Just like Physx hasn't changed gameplay and is only used for sparkles and stuff, Mantle isn't going to change the game and is just going to be used for a very slight performance boost. Maybe a major performance boost in the extreme case of RTS.
Which is a shame. But that's what happens when you go the vendor-locked way.
Which also means you have no idea what you're talking about.
What can more likely happen is, Nvidia losing ~30% market share by the time they get to Denver offload CPU (if that's what's going to be anyways)
Vendor lock-ins and proprietary tech only fails when they don't have a compelling advantage - that's why PhysX was a failure, it just didn't matter enough to make the AMD buyers switch to Nvidia.
Exactly, all the signs point that Mantle is going to outweight the strength of brand loyalty.
So, if Denver Offload CPU works 10 times better than Mantle, so be it, but since that's still a rumor why should I go even bother going deeper into that debate right now.
Mantle uses HLSL to write shaders, the same as DirectX- it's not like developers are going to be writing GCN assembly.
The more "low level" control is stuff like memory management, and more explicit control of queues and scheduling. EDIT: Or at least, that's my interpretation as an uninformed non-game developer.
Emm, actually it's a HLSL converter afaik. They probably can start from scratch if they want.
Really. Since you weren't around back then and make that completely obvious, let me clue you in. Glide was created back when D3D didn't exist with the Voodoo I. There was NO standard, there were 11 standards all competing with each other - every chip had their own 3d API. PowerVR had their own thing going on. So did the rendition verite. In fact, when D3D became ubiquitous there was no longer a need to get a 3dfx card. Eventually 3DFX fell so behind on the D3D performance curve, this with other factors (V3 was a disappointment and late, purchase of STB, etc) led to their demise.
Thank you, drive through. Oh yeah. Speaking of which. Most glide titles did not have a big performance advantage over D3D. It was generally 10 frames if that - which didn't matter when 3dfx fell behind the curve in terms of D3D performance. The voodoo 3 was a disappointment but they barely managed to hang on. Let's see, from what I remember, the Riva 128 was the competing product to the voodoo2. The V2 was still better in terms of D3D performance, and very few titles used glide by that time. Then the Riva TNT was released - it was just better than the V3 by a mile. The V3 was also late and disappointing, with poor D3D performance. Shortly after, 3dfx went caput.
3DFX initial success had nothing to do with glide. They were successful because...well...mostly they were one of the first and they had the highest performing overall part by a mile. Nobody could touch 3dfx performance by any metric. 3dfx was loved initially because they got in on the market early and they had the most compelling performance and hardware. Once that advantage went away, people stopped caring. It had absolutely nothing to do with glide, like I said glide was created because there was no standard in the early days. It was proprietary or nothing.
That's all fine, but. DirectX wasn't designed in the year 2013, It was designed in late 90ies with code being piled ontop of that methodology. Similar with OGL.
And now, we don't have 11 APIs, we have 2, does OGL even count in the AAA PC market ? (i know it's for on Mac versions and exceptions like Rage etc)
AMD with Mantle, with avoid all the fuss huss when they had with Rage release (ofcourse if DX fallback path is provided which is definitely yes for quite some time, they'll still be prone to those problems on that version)
when the wrong driver got released and game basically didn't work, worse yet, OGL drivers were quite bad from AMD so it didn't work that great even after some weeks, eventually I played it like a year later and it worked great. But see, AMD can't fix the driver good if the developer doesn't cooperate enough, which Carmack admitted they focused too much on Nvidia (OGL extensions), so without that communication going
With Mantle, developer doesn't need communication, he can fix it him self, exactly what Carmack could do for rage to get it working and there would be no issues like that, because he would took the time until it's done right, from my research I see they had a really clean codebase that was scanned by a static analysis company and they got the tools and basically polished to that extend they were treating warnings as errors.
For comparison, Rage was basically all graphics bugs, while BF4 is mostly all in gameplay. So it's true Mantle won't fix and won't do anything if games are rushed and the gameplay side doesn't work it doesn't matter if there's mantle or not, but, by being so much easier with mantle, they probably won't have to spend that much effort on optimizations, and I think that could lead to better overal quality.
3) DX was used by M$ as their weapon to get Windows in to everybody's homes. It's no longer the case and hasn't been for the past few years. They simply don't care enough - do you actually believe Microsoft is wasting time on DirectX these days while they're being owned all over the shop in mobile?
I also lean more to DX primairly being a weaponized API rather than "solving the issues of the industry", that compatability/security is just a smoke screen, and in year 2013 that is more obvious than ever.
Sorry I care about gaming...as in the graphics (I.Q), the framerates, the gameplay and the control-fidelity.
I eg. game ARMA3 in 7.1 surround sound with TrackIR 5.
No crapbox can do that.
Consoles are for people not serious about gaming.
Allright. But more to those argument, we try to defend the PC tradition and culture from being dumbed down and sabotaged by outsiders or potentially paid agents.
Mantle will improve many of those criticism of PC being too hard to tweak for everyone, and constant driver thinkering, well there's no thinkering taken away, but there are some serious annoyances that will be obsolete if you go Mantle mode.
If you don't know how to thinker with settings then that's not a problem of a platform in the sense that the whole platform should change for everyone to be forced onto the easier method. Mantle will fix the genuine criticism, which is caused by high-level APIs.
But I love to setup my game settings for hours before I play, and I absolutely refused to play games without my custom settings, for example Crysis one of the big PC favourites that is at the top of the list with Zelda and Metroid, I created my own autoexec.cfg and setup about 100 console commands I wanted the game to look, especially draw distance, I extended it by more than double, maximizing audio quality settings, sorround, etc
There's a reason people with high end PC's buy consoles.
You can claim "I can get 1000 fps" but what's it matter if it drops to 1 fps, or suffers from periodic or intermittent stuttering? Console games have no issues because game devs have control over the hardware, and that's what AMD is trying to get on the PC.
Me, I don't buy consoles for "easy easy" and "no issues", I buy it for Zelda and Metroid (and I hope they finally make another F-Zero or Starfox).:sneaky:
Soul Calibur switch wasn't enough for me to get a X360. Especially the consensus being that the Soul Calibur 2 on GCN was still the fan favourite, but idk, haven't looked into that much lately.
ArmA 3 and RTS games like Total War would probably benefit from Mantle more then a gpu-limited BF4 would.
Indeed
Heh a minority player in the GPU market is going to push a non-standard API and make it the dominant standard? Do you guys read what you are writing?
Well what really is the point, is it criminal for an underdog to do that ?
Alternative media was also a minority against the mammoth mainstream dinosaur media, but now it's the other way around.
DX is high level and provides high compatibility.
Something I can, personally, throw down the toilet.
Acceptable Sacrifice.
DirectX and OpenGL serve a purpose,
Something
nobody cares about anymore:
Year 2011:
http://www.bit-tech.net/hardware/graphics/2011/03/16/farewell-to-directx/
Huddy says that one of the most common requests he gets from game developers is: 'Make the API go away.'