it depends on the source of this. first actually i was expecting 4 is nonsense 5th is epic fail if true rest are not that important imo maybe the portrait thing.
Printable View
it depends on the source of this. first actually i was expecting 4 is nonsense 5th is epic fail if true rest are not that important imo maybe the portrait thing.
Nvidia would rather makes something of their own(not the low gain Nvapi or whatever it is) and make it exclusive.
If and when Mantle opens up a bit more, the primary benefit in my mind is to increase attractiveness of AMD CPU's. The gains with generally poorly performing AMD CPU's mixed with about any GPU appears where most of the gains are.
The performance increases of the high end don't appear that impressive at all. If AMD was to make Mantle purely closed for AMD GPU's, it would hurt the attractiveness of their CPU's as Nvidia still commands the greater market share.
One thing I think would help amd but I question if 3dmark would allow it since it introduces some unfairness for Intel is to redo, 3dmarks or do future versions in mantle. Currently, no one uses amd processors with 3dmark to break records. They are simply awful. With a mantle version, this could change in a big way.
Yes! I have an elephant in scotland AMD yield figures over their last year's earnings since Intel is releasing 14nm double waffles made out of LN2 with high clockable ability.... god I haven't even written one line of crap and I'm already at my limit... do not know how you can post relentless useless info the way you do, but I gotta hand it to you you write some of the best useless crap out there.
^^ his posts are like Osmium.... dense as Fcuk
Rofl!!!
it needs to be an epic fail for those who were expecting 30-50 % straight off the bat whilst on Beta drivers... it makes sense... after all we're used to seeing 20 million thousand percent increase from gen to gen..... when buying videocards for free too (not sure if I was cynical enough there)
here are some facts.
API coding takes time, and is highly dependant on a multitude of factors
this is the first time we see "coding to the metal" in game applications
BF4 has a heck of a lot of issues that still need sorting
anyone who calls this an epic fail is proving to know nothing on the subject
same, but a bad publisher can turn a nice game into crap rather fast
yeah I used to be in -AOS- for Battlefield 2 & 3
few times sitting in the top 5 AUS clan.... we had 3-4 kiwis too, their ping wasnt too high and they were mad
Johan Andersson article about Battlefield 4 AMD Mantle, good read. Lets see what tajoh thinks about it, I'm ready for another good read.
This part is interesting, just to show Mantle is not ONLY about AMD CPUs
Read more hereQuote:
Test case 3: High-end single-player with multiple GPUs
CPU: Intel Core i7-3970x Extreme, 12 logical cores @ 3.5 GHz
GPU: 2x AMD Radeon R9 290x 4 GB
Settings: 1080p ULTRA 4x MSAA
OS: Windows 8 64-bit
Level: South China Sea ?Broken Flight Deck?
This single-player scene is heavy on both the CPU and GPU with lots of action going on. Test was done on the highest end Intel CPU on Windows 8, which is the fastest option before Mantle thanks to DirectX 11.1. Still this CPU is not fast enough to keep the 2 290x GPUs fed at 1080p on Ultra settings so we get a significant CPU performance bottleneck which results in major performance improvement when enabling Mantle.
Result: 13.24 ms/f -> = 8.38 ms/f = 58% faster
Exactly point out what is exactly wrong. Please do an explanation that shows some effort rather than douche comment.
Really add some graphs and stuff as well while your at it and show something beyond this snarky attitude. My post isn't even negative towards AMD.
My first comment was aimed more negatively at Nvidia than anything. Nvidia is generally pretty close minded with all their software because their greed keeps this software features proprietary.
And for the rest of my post...
http://www.hardwarecanucks.com/forum...es-mantle.html
http://i62.tinypic.com/5y6ye1.jpg
http://i62.tinypic.com/33nvign.jpg
http://www.extremetech.com/gaming/17...pus-not-gpus/2
In Battlefield 4, we have more robust information. AMD claims that the FX-8350 + R9 290X will see an average of 23.8% better performance under Mantle, while the Core i5-4670K + 290X will see a performance improvement of just 7.5%.
Look where the biggest gains are. They are with lower end AMD APU's or AMD CPU's in general.
If you have a mid to high end Intel CPU paired with a high end GPU, Mantle isn't exactly boosting much thus as I said earlier if you look at the above graph, Mantle isn't aimed at boosting the performance of the enthusiast so much with a high end GPU. However add mantle to a slower AMD processors and you get pretty good results(quite fantastic in some cases).
With the small adoption of mantle currently, it acts as little leverage for people to buy an AMD CPU or even an AMD GPU(unless they have an AMD APU paired with an AMD GPU). A person is better off with an Intel processor because they get consistent results across all games(not just Mantle ones).
If someone gets a new computer and it comes with a decent Intel processors, they currently get good performance with either an Nvidia GPU or AMD GPU. Currently without Mantle, getting an AMD APU, mixed with a GPU from either Nvidia or AMD, you get something something thats between mediocre to almost matching Intel(atleast with an FX, less so with an AMD APU). With Mantle, your finally able to get good performance out of some games(actually one only) with an AMD APU. Limiting it to AMD GPU's would make the Intel solution more attractive again if it wasn't already more attractive.
Getting an Intel CPU currently nets you the following.
Good performance with AMD or Nvidia GPU's.
Good performance consistently across all games.
A slight boost with Mantle.
Getting an AMD APU currently gets you without Mantle.
A mix of performance with AMD or Nvidia GPU's.
A mix of performance consistently across multiple games depending on how CPU limited they are.
This is what made people not want to mix discrete with AMD.
With Mantle, AMD gets the following benefits.
AMD nudges between 2 to 10 percent more performance in games that support it when combined with a high end AMD GPU with a High end Intel processor
A 30-50 percent boost when combining an AMD GPU with an AMD APU(which still performs poorer than the Intel result).
But most importantly, you only get the above benefits when the game supports Mantle. This makes or breaks the benefit of mantle, thus mass adoption from developers is what AMD needs to make it a clear selling point. To get mass adoption without a cash payment from AMD, its going to require both support of AMD and Nvidia in developers eyes. From the time it took to implement the mantle version of BF4, the work is not trivial.
At this point it should be clear the benefit of Mantle isn't so much for AMD GPU's as much as it is for AMD CPU's. The above article I mentioned even titled it something along these lines. And this is exactly what AMD needs, a bigger reason to use their CPU's in general. AMD's CPU division has the most room for revenue growth and it is currently the poorest performing area for them. Thus to keep Mantle purely closed for AMD GPU product, would limits its effect of boosting CPU sales because ultimately the Intel processor gets more consistent and higher performance across the board and it doesn't need Mantle to do this. It would hurt its adoption from developers and people that currently own an Nvidia GPU would have no interest switching to an AMD CPU.
The only way to get 58% more gain is if cross fire was broken in the first place.
100% scaling is pretty much the ideal and impossible to get.
If initial performance was say 1.8(80% percent crossfire scaling) vs a single 290x, that would mean with a 58% improvement(1.8x1.58), 284%(or 184% crossfire scaling) improvement over a single card was achieved and that is impossible. However say performance was 1.2(or pretty much broken crossfire), a 58% improvement would be 1.896(roughly 90 percent crossfire scalling which would be fantastic still). Thus I suspect this is more of a fix for broken crossfire scaling.
essay?
Then Mantle would really benefit AMD in the low power segment such as a laptop or other portable device that relies on an AMD CPU/GPU combo. In that scenario we could possibly see AMD yield better performance per watt drawn with Mantle?
damn, that Johan guy has absolutely no idea what is he testing and the worst part of it he must run it at least couple times just to make sure before publishing. Maybe you should apply for a job there since you know so much better.:) You ask for benchmarks and when i show you one you still not happy.
I don't want to dispute Mantle at this time might benefit mostly mid or lower end setups but it looks like ONLY it's not the case.
Anyway unlike you I have very hard time to find anything wrong about it.
If you have high end cpu and dont like Mantle there is one option you can do:
Dont use mantle. That will show them.
People with low end cpus or apus propably dont mind getting free 0-40% performance increases in some gaming scenarios.
edit. This is bit offtopic but basically similar message in 0:20-0:50 https://www.youtube.com/watch?v=dqXWS7Efgcc
Hold up, why is hardwarecanucks ignoring these numbers? From PCPer's article.
Core i7-4960X CPU + R9 290X GPU
1080p, Ultra Preset, 4xAA: 9.2% improvement with Mantle
1600p, Ultra Preset, 4xAA: 10% improvement with Mantle
10% increase is a good amount of increase.
yes at that level %10 is amazing
Hi guys... long time no posting anything hehe...
http://www.guru3d.com/articles_pages...preview,1.html
Preview by Guru3D
For what I'm seeing, the problem with CFX & other implementations in BF4, it's not problem of Mantle, but a problem on the game side itself... Maybe with Thief we will see a much complete view of how Mantle benefits the CPU & GPU in games (sorry if my english it's a little complicated to understand... spanish is my native language hehe)Quote:
Knows issues with the current Mantle driver build:
Now I am not too interested in the issues as they will be solved and this is a Beta driver, but it does tell the current state of what current product are actually supported, have a peek.
Press evaluation build: Installing AMD Catalyst 14.1 Beta on top of an existing AMD Catalyst revision may cause a system hang. Ensure the provided installation instructions are utilized to avoid this issue.
Intermittent stuttering or stability issues may occur when utilizing Mantle with AMD CrossFire technology in BattleField 4
Mantle performance for the AMD Radeon HD 7000/HD 8000 Series GPUs and AMD Radeon R9 280X and R9 270X GPUs will be optimized for BattleField 4 in future AMD Catalyst releases. These products will see limited gains in BattleField 4 and AMD is currently investigating optimizations for them.
Multi-GPU support under DirectX and Mantle will be added to StarSwarm in a future application patch
Notebooks based on AMD Enduro or PowerXpress technologies are currently not supported by the Mantle codepath in Battlefield 4™
AMD Eyefinity configurations utilizing portrait display orientations are currently not supported by the Mantle codepath in Battlefield 4™
Graphics hardware in the AMD A10-7850K and A10-7700K may override the presence of a discrete GPU under the Mantle codepath in Battlefield 4
AMD testing for the AMD Catalyst 14.1 Beta has been concentrated on the following products: AMD Radeon R9 290X, R9 290, R9 280, R9 270, R7 260X, R7 260, HD 7000 Series, HD 8000 Series, A10-7850K and A10-7700K. Future AMD Catalyst™ releases will include full test coverage for all AMD products supported by Mantle.
Concluding
This is an extremely short article as there is just so much to explore with mantle. It will remains to bee very simple though. On high-end PCs with a single graphics card you may see perf increase in the 10% area. On massively CPU limited situations like low/medium quality settings / low resolutions or even Multi-GPU setup, that's where the gain easily can get 30 even 40%. As such Mantle will benefit low-end and mainstream processor based PCs the most. But even at Enthusiast class, it will make a difference.
So would we use Mantle ? Sure .. any free gain is a plus in my book. On the next page a FAQ published by AMD to address some questions you guys might have.
Looks like the stuttering issues they were talking about is causing spikes in frametime thus raising frametimes in a high-end scenario, although you can see an overall increase in Average FPS.
http://www.hardcoreware.net/battlefi...log+(HCW+Feed)
I like what Guru3D is saying because I play BF4 MP with a mix of Med and High settings to get higher FPS and thus making myself CPU bound with my Crossfire setup...if they do optimize for 7000 series in the future then awesome...I will accept a free performance upgrade!
ROFL thanx all the way from Brissie.
I swear someone ought to limit his post word count to like 100 words.
that guy is becoming more annoying than Terrance123..... other than that his "larger than life, denser than a red star" posts.
now on something new, I think for those who have AMD cards to post mantle benchies with different configs once the 14.x drivers are out
I have an 2600K (stock right now... it's my gaming/mining cpu at this time... but can be overclocked to 4.5 ghz) + R9 270X Windforce OC 2Gb + Star Swarm Demo... I could post my results at stock & overclock once the driver is out, to see the diference (I only have Fraps to see the FPS on BF4... are other options to see frames or frametime available?)
http://www.golem.de/news/amds-mantle...-104261-3.html
64 player benchmark. Also, 43% increase in min fps... holy crap. Need more data.
http://i.imgur.com/0ilDBlb.png
There's a bunch of Star Swarm videos on the net. Unfortunately, none on Mantle, yet, except for those by Oxide themselves. I would watch them myself but it seems like my drivers can't run it (probably needs 13.12 WHQL. Mine is 13.11 beta).. and I simply can't be bothered to update when the Mantle update is right around the corner.
http://www.youtube.com/results?searc...week&lclk=week
They're nice to watch.
go on guys. amd will not release drivers until we discuss all the possible things on mantle :D
Why I can't download Mantle ? January is over now, no ?
AMD seems to have a serious problems with communication/execution.
Anyway ...
http://tof.canardpc.com/preview2/3b4...58ba72c44a.jpg http://tof.canardpc.com/preview2/67d...efe90c9a69.jpg
http://tof.canardpc.com/preview2/158...d2e62dbe61.jpg http://tof.canardpc.com/preview2/b52...bb59ffd5ce.jpg
http://tof.canardpc.com/preview2/994...0d28346000.jpg
http://pclab.pl/art55953.html
/\
||
Somewhat cherrypicked results, here are other ones:
http://pclab.pl/zdjecia/artykuly/cha..._a10_7850k.png
http://pclab.pl/zdjecia/artykuly/cha...pu_i3_4330.png
http://pclab.pl/zdjecia/artykuly/cha..._cpu_4770k.png
http://pclab.pl/zdjecia/artykuly/cha..._cpu_4670k.png
And the one that interests me the most
http://pclab.pl/zdjecia/artykuly/cha...cpu_fx8350.png
Outstanding.
Where CPU isnt any bottleneck at all, theres little to some difference, when theres a signifcant amount, its night and day world of a diff
It literally DOUBLES 7850K performance in heavy MP scenario, it turns this cpu from not recommended to recommended.
And turns my aging 8350 from a dog with broken leg, to something that catches up to new haswells.But hey, im just waiting for somebody to tell ME that mantle "sucks" and "epic fail"
even cherrypicked results are awesome. very good gains over directx.
How did they test multiplayer in pclab.pl? Did they ever mention it? The previous site I linked (the german site) said they got onto a boat, drove it to a bridge, average 5 runs, and throw out those that vary wildly (google translation might be sketchy). pclab.pl, nada.
Apparently drivers not out means bad communication to Olivon.
+1.5% in SP is quite deceptive.
Hopefully, MP is much better with almost 24%
Even if it' doesn't beat a 1200MHz 780 in this test in MP mode, the gain is really noticeable, even with an i7-4770K stock or oced.
Dude, how is that even the smallest gain that you cherrypicked "deceptive" ? And of course the lower performing cpu you get the gain is goin to get bigger in SP too, so deceptive is calling it overall 1.5% gain, cos thats probably the smallest one you can get.
Deceptive is the way you talk about the results in MP also.They range from 24% to a 100% gain in MP,and 70% for a FX 8350!.If you want to call exact results than say 24% gain using 4770K cpu.
What i see is that this thing, is gonna be HARD to benchmark, depending on the cpu/ram/gpu and a scene used in a game, it can probably vary between 0 and 100%
I was talking about 4770-K because it represents the best choice for pairing a 290X class card.
Even the i3-4330 got slightly better results than FX-8350 in this review.
But I tend to agree with you about the difficulty to have accurate results in MP mode.
Better move to do is to wait from other propers reviews to come.
Apparently you dont understand that what it means is that in mantle enabled titles it IS NOT a bad choice to pair up 83XX with 290X.It means exactly that i can buy 290 and still get very good results.When money is tight people can save money and get weaker CPU and put saved money towards a faster GPU.I was kinda on the fence about going haswell , but now, it may not be needed.
While that is true, it is only true with Mantle titles, like you mentioned.
While Mantle seems to have a decent backing of titles rolling out over the next 6-12 months, all the games without Mantle support will be at a disadvantage with certain "slow" CPUs.
Obviously this disadvantage can widely vary, from noticeable to not noticeable, and taking myself as an example of an adamant PC gamer a large backlog of games where a newer better performing CPU may not have a large affect on performance, it still needs to be accounted for.
It will be interesting to see what gains can be accomplished as Mantle matures in regards to pure highend GPU performance.
I agree fully, its not like im recommending people to build brand new Am3+ systems.But considering not every title (in reality most of them) are really that much cpu limited,and there seems to be considerable mantle backing it all looks good to owners of Phenoms X6 ,FXes, first nehalems core2quads etc.
Its just that Olivon seems to be putting things into worst perspective possible, not everyone owns 4770K or IVY-E! And people who do, can have fantastic CF scaling!
You cant argue with worst case scenario as if this was applying to everyone.
One thing is sure, mantle has big potential.
We are on Xtremesystems dude.
If you can buy a 600$ card, 100$ more on the CPU is not really a problem.
Don't forget, it's just the debut of Mantle, with only one game (almost) available.
I dunno for you, but I don't buy a CPU or GPU for only one (buggy) game.
edit :
http://tof.canardpc.com/preview2/e35...93a43f9256.jpg http://tof.canardpc.com/preview2/b4d...0abf0e1ae0.jpg
With Mantle activated, the i5-4570 do better than the FX-8350 oced to 5GHz.
If you count the cost for competitive cooling + strong power supply and motherboard + energy cost, FX-8350 is not really attractive according to this review's numbers.
edit 2 : AbortRetryFail?, Thanks for your constructive comment. Always appreciated on XS newz section those days.
Taj ?
Is that you ??
All I gots to say: Yah can't beat free.
:)
So what, if we are at xtremesystems, and thats your "argument" shouldnt we talk only about 290X trifire with i7 4980X ? (which would get big gains on mantle btw)
Or was that dumb argument right there ?
About "having money" and ways to spend it, its not that easy as you make it to be, if you dont have to buy new cpu you can invest int more gpu power, and thats where mantle does good, and in effect you get more performance for less money.Also there are many problems building pc with a fixed amount of money,and no, not everybody thinks 100 or 180$ (thats the diff between 8350 and 4770K here) is "nothing".
Also, 290 and 280X for example are a lot cheaper, so another DRAMATIC argument there.
Of course nobody buys cpu or gpu for one game, but this thread is about mantle and not BF4, BF4 is just first beta iteration of it so thats why we are talking about it now.There are more coming.
You cant pick only situations that match your agenda and be like "Thats like things are globally".Not gonna respond more to arguments like this.
TO Olivion.
As for your link about 8350 oced to 5ghz and whatnot, you gave the link to DX version, heres mantle:
http://pclab.pl/zdjecia/artykuly/cha...eon_mantle.pnghttp://pclab.pl/zdjecia/artykuly/cha..._radeon_dx.png
EDIT: Edited to have them side by side, MANTLE - DX
Also im not saying people should be buying FXes now for christ sakes, im saying they can buy cheaper intels and amd`s or just stay with what they got.
AMD Passes On Catalyst 14.1 beta to the Press, Public Release Shortly
Read more at TPU
No problem man, im sorry if you felt that way.They updated the Mantle Cpu graph, with old grandpa Phenom X4, as you see on mantle it pulls out some respectable numbers!
Just to put things into perspective, Phenom X4 is doing on mantle almost the same FPS as i7 4770K is doing without.
Short preview at guru3D
http://s3.postimg.org/jvqkmjer7/untitled_3.png
http://s29.postimg.org/hb50pba6f/untitled_4.png
http://s29.postimg.org/dx0meeix3/untitled_1.png
http://s28.postimg.org/6sbyqam2l/untitled_2.png
Quote:
Concluding
This is an extremely short article as there is just so much to explore with mantle. It will remains to bee very simple though. On high-end PCs with a single graphics card you may see perf increase in the 10% area. On massively CPU limited situations like low/medium quality settings / low resolutions or even Multi-GPU setup, that's where the gain easily can get 30 even 40%. As such Mantle will benefit low-end and mainstream processor based PCs the most. But even at Enthusiast class, it will make a difference.
You should see some very nice increase in performance with your Crossfire 290x's. It might make me look at getting a second card.
I just got my two Sapphire R9 290 4GB TRI-X OC couple days ago, looks like Newegg.ca approved more orders than they had in stock and I had to wait for new shipment.
I had some time yesterday to install the card also with couple new case fans. When I started the PC there was some bad fan noise, disconect the new fans and noise was still there. Removed one card and the noise was gone, replaced the card and also not any noise. Installed them again in crossfire and the noise was back. Figured out the cards were sandwitsh too close together, when I wedged them a little the noise was done.
Removed the second card, didn't want to damage anything and ordered new board GIGABYTE GA-X79-UP4 LGA 2011 plus new CPU Core i7-4930K and G.SKILL Trident X Series 32GB (4 x 8GB) DDR3, everything should be ready by Thursday next week.
In the meantime the one card works great but I didn't have much time for testing any OC.
Sorry guys about this off subject post.
LegitReviews
AMD Mantle API Real World BF4 Benchmark Performance On Catalyst 14.1 (Feb,1st 2014)
The difference in cpu limited circumstances is pretty damn impressive.
Makes the PS4/XB1 cpus seem much more capable. Seems so far brute forcing still nets solid results but that fact that they are doing twice as much with half the cpu power is great.
Be curious on the kind of gains a MMO could see with this kind of treatment (alot of them are stupidly cpu limited) I have a feeling the underlying engine itself will be the biggest element for improvement with any game though. Frostbite is fairly forward looking and modern compared to some more antiquated engines (ie ones which still render in DX9 and dont scale with multiple cores) Have high hopes what Mantle can do for Unreal 4.
lets see how the Tajoh focus group will talk this one down.
edit** if AMD gets enough devs on board and optimised enough AMD will welcome Maxwell with open arms like an orphan child "come here little one... see that? it's mantle, also known as your kryptonite"
lol
but on a serious note, this will turn the tables on a huge way for AMD, throw in some HSA and HUMA.......droool...
My new HTPC/LAN box is going to enjoy this :D
-PB
We need to see the results in the low-power segment (e.g. laptops and gaming tablets)...that is, is AMD's CPU/GPU on one die netting more performance per-watt when using Mantle as opposed to an Intel/Nvidia combo or an Intel/AMD combo. This is where it might really shine...does nobody else think this?
Found a link to the 14.1 drivers.
http://support.amd.com/en-us/downloa...os=Windows%208
http://support.amd.com/de-de/downloa...os=Windows%208
Links are going up everywhere now
So far so good. Battlefield seems a bit more smooth. Frames are fine without any noticeable stutter.
downloading now:D
Bahhh, I'll wait for Win7 64-bit drivers. I'm not sure if they're the same. EDIT: Nevermind, I'm pretty damn sure it will work with Win7. There's no 'win8' label on the driver itself.
It seems after my initial great first up experience with Mantle enabled, I now get a very slight stutter in BF4 every so often. Happens maybe once every few minutes. It's not constant as on the whole, it appears to be smoother than with the 13.12 drivers. Still it is enough to get you KIA in Battlefield.
I think my PC deserves the windows update it has been wanting (been sitting just behind firewalls and has been fine for, ugh, 5 years). I'm still getting grey screen on the Star Swarm demo,
Funny how a Star Swarm FAQ page is required to answer all the statement saying AMD bias on Star Swarm.
http://www.oxidegames.com/2014/01/31/star-swarm-faq/
I would get an unrecognized display driver error after installing 14.1 on win-7 64bit when both 7970s were connected. So I had to disconnect power from one of my 7970s, boot into windows, install the driver, reboot and let the driver successfully load for the first time....then I shut down, connected card 2, at booted back into windows and the driver now loaded successfully with crossfire enabled. Maybe this will help someone...
give us some feedback if you can on in-game experience.
If you were asking me, sorry but I use crossfire and it states in the release notes that crossfire is borked for the time being and I sincerely don't want to play with a stutter...and I'm lazy and that's why i'm here right now is trying to get other peoples experiences:)
Edit: actually I will go try it out right now and see how 7970 crossfired + Mantle does...brb.
Ok, in single-player with the 7970s there is no stuttering which is good:
I benchmarked the Framtimes of the Jeep ride in the beginning of 'Tashgar" and my frame-times where between 3.44 and 7.62ms so no spikes,
but also no noticeable performance boost...yet (DX api yields the same fps as Mantle api). My settings are a mix of Low, Med, High, and Ultra...so yes I am intentionally leaning towards a CPU bottleneck to raise my FPS, so I would be-able to clearly see a benifit if I were using Hawaii GPUs but as stated the 7000 series is not yet 'fully' optimized so as expected nothing here.
I hope they optimize 7000 series like they say they will...
On my 290x I get a strange hitching every now and again. It's not constant, but is annoying. I tried uninstalling and reinstalling the driver, but same problem. I do have a custom user.cfg in BF, so I will remove it and see if it helps.
Performance is great. On my 27" 120Hz 1080p, all settings maxed, I get high 80's, without Mantle I was in the high 70's. Without any AA modes, i'm always over a constant 120FPS.
we need to know your CPU as well MaddMuppet.
I'm still getting grey screen with Star Swarm. Seems to be related to HD 7850, since all the ones reporting the bug is on 7850. EDIT: Not only on 7850.
Anyone with Star Swarm working could help me follow these steps and show me what redistributables you have installed?
Control Panel > Programs > Uninstall a program > Sort by name by clicking the Name tab > Scroll to Microsoft Visual C++ > Take a screenshot of all the redists with the version numbers. Thanks.
I have 14.1 beta drivers installed and ran some tests of starswarm . 7970 BE @ stock speed 15% powertune :
standard settings :
http://www.skybed.be/temp/xtremesyst...SW_Directx.jpg http://www.skybed.be/temp/xtremesyst.../SW_Mantle.jpg
custom settings (2560x1400 , NO BLUR , Glare 48 , Shading Samples 160)
http://www.skybed.be/temp/xtremesyst...ctx_noblur.jpg http://www.skybed.be/temp/xtremesyst...tle_noblur.jpg
@blindbox :
Microsoft Visual C++ numbers
Thanks! This is most helpful. I noticed I don't have some of those.
got 3 rounds with mantle with my 7970 lightning. seems not to have a fps increase but if there is nothing special about this 3 rounds it feels smoother.
^^ im on dual 7000 series too, so we'll see what eventuates..
i have 2 x 7870 and 1 x 7970... ill see which setup works best atm
Can anyone do a picture comparison? I have seen examples on other forums where Mantle reduces the quality a lot.
It's supposed to be a bug, makes some places too bright. People who have taken the pictures to Photoshop reports that the objects are still there, just too bright.
Btw, if anyone has issue with going to "Options" in BF4 crashing the game, someone in battlelog forum found that it has to do with having Intel integrated graphics (in addition to a Radeon), it has to be disabled in Windows device manager.
why am I not surprised that this is full of bugs as usual for any dice patch...:shrug:
Did any other sites do any meaningful array of tests ? Anandtech is pretty much a joke ,guru3d isnt too comprehensive also.
Anyhow two shots of task manager DX and MANTLE which explains somewhat higher scores using mantle:
http://pclab.pl/zdjecia/artykuly/cha...irectx_big.png http://pclab.pl/zdjecia/artykuly/cha...mantle_big.png
CPU usage goes considerably UP.
So has anyone else had a slight pause or stutter every few minutes in battlefield. I'm trying to work ouy if it's just my PC or a beta driver issue. I'm on windows 8.1 and it's a fresh copy.
Ok I installed it last night on my current rig which is 3930k@4.6GHz and 290x @1080/1500
As far as I can see my fps is pretty much the same as before... And it might be smoother but really I couldn't tell the difference lol
Looks like it's a common issue. Still it is only a beta version.
http://battlelog.battlefield.com/bf4...24388458174/8/
I'm holding back until the official driver for the moment. I've never been too keen on beta drivers except in rare cases.
Sorry double post, but I agree with you completely Sparky...AMD just wanted us to see that the benifits of Mantle are very real and will certainly bring benefits if supported by developers.
Take these beta drivers for what they are...A Technology Demonstration and we are the Beta testers :).
This may sound dumb, but sometimes we miss the simplist of things. You did close down the game and restart it after enabling Mantle. It's just that your system and mine are fairly close in terms of hardware and I did notice a difference in performance and smoothness. (Except for the odd stutter or pause)
err..thinking about it now, i dont think i did, most of the video settings are set on the fly, so i just changed the renderer, then started playing, during playing i was toying around with AA, 2x and 4x to see if I could find any performance difference since those settings were making a difference I assumed that the renderer setting was also set properly.
You are correct though I should have restarted the game, I was pretty peeved with the CRU utility which I use to add custom refresh rates (which I have to do every driver update, because atimagwhatever needs to be patched) because it took me 3 restarts to get the custom refresh rates in. I will try it again tonight after i get back to work. thanks for that
I get absolutely no stutter in single player...I even used 'perfOverlay.FrameFileLogEnable' to output my frame-times for about 5 minutes and I never had a spike in frame-time. Now, that's with crossfired 7970s and obviously they did all the tweaks for the Hawaii GPU so there might be something there.
Best way to find out is to play Single player and enable the "frame file log" and see if you have any spikes in frame-time. Do it for both DX and Mantle to see if the Mantle spikes are consistent with DX, if not then new anomaly has been introduced by Mantle.
One other issue i'm currently having is I can't get any OSD to display in BF4 with Mantle. No Radeon Pro or Afterburner. I see others don't have this issue, so it must be related to me. I am running Battlefield in 64bit mode, if anyone else gets this.