Nvidia vs AMD drivers

[IMPERIUS]

Hey guys,

I've been hearing a lot of bad stuff about nvidia drivers latelay, the biggest problem being a 960 beating a 780 in the witcher 3 and just general problems all around. But AMD carries this myth of bad drivers his whole life. Personally I haven't had any problems with them.

I'd love to hear your thought on the matter

[Deletive]

Upgraded because I couldn't stand the daily crashes on my 750ti (unoverclocked) driver would always crash everywhere, on multiple versions. Switched the AMD and has crashed once since January (when I upgraded) 

[Vavetti]

My amd driver (using 280x) only crashed once during gaming in over a year of use.

[Hunched]

Nvidia has been breaking everything software related they touch for like the past year so...

[Raffy98]

I have an R9 270X, AMD Drivers never crashed in a year since i have this card!

I don't get why people say bad stuff about AMD drivers.

[Kinda Bottlenecked]

Both are equally shitty, neither is better. 

[Dabombinable]

Its more like who's best at fixing the mistakes of game devs, and it really depends on the game.

[ybriK]

Both drivers have had their bad moments, no driver is better than the other look back at ATI, now look at Windows 10. How to choose a graphics card when short on money? Best Price to Performance Ratio.

[NinjaTuna]

Nvidia drivers has hit that phase of turning into pile of crap; It stopped supporting bandicam's Nvidia plugin, crashes more often and so on. I was forced to roll back to older versions.

[Majestic]

Hey guys,

I've been hearing a lot of bad stuff about nvidia drivers latelay, the biggest problem being a 960 beating a 780 in the witcher 3 and just general problems all around. But AMD carries this myth of bad drivers his whole life. Personally I haven't had any problems with them.

I'd love to hear your thought on the matter

 

I'm not going to argue in the subjective and anectodical "driver never crashed for me, must be no problems for everyone".

 

My issue with the AMD drivers is the poor performance it has on DX11. And is partly to blame for the reasons why they perform so poorly in those gameworks titles. It's not because nvidia's gameworks implementations are black boxes, it's because they put even more strain on the CPU and that is a problem since AMD drivers already strain the CPU more. As most of the gameworks features, and phsyx, are run on the CPU and don't have GPU acceleration.

 

Look at these results.

 

You can clearly see the Nvidia drivers getting almost double the amount of drawcalls out of the DX11 API. Meaning they're much more efficient than AMD's DX11 drivers and have less overhead.

 

The difficulty with this is, GPU testing is done with fast CPU's. Usually 4790K's or 5960X's running overclocks to remove any form of CPU bottleneck. But that also means they have way more capacity to deal with inefficient drivers than say... locked i5's, i3's or AMD FX CPU's. Meaning the R9-380 which, suggested by this forum all the time, does so well vs. the GTX 960...doesn't acually perform that well in the budget build it's being placed in. And that pairing a R9-380 with a 'lowly' i3-4160 or FX-6300 might not provide better results as the same CPU's + gtx960. I say lowly for the sake of argument, because next to the 5960X @ 4.5ghz, the 4160 is kinda lowly (even a 4460 is) 

 

 

An example of a CPU heavy game. Normally the GTX 980 and R9-290 wouldn't be this far apart. Neither would the GTX 770 and R9-290 trade blows. Yet now they are, running the same CPU's.

 

EDIT;

 

The latest DX11 vs. DX12 tests shed some light on what i've been saying;

 

 

[IMPERIUS]

I'm not going to argue in the subjective and anectodical "driver never crashed for me, must be no problems for everyone".

 

My issue with the AMD drivers is the poor performance it has on DX11. And is partly to blame for the reasons why they perform so poorly in those gameworks titles. It's not because nvidia's gameworks implementations are black boxes, it's because they put even more strain on the CPU and that is a problem since AMD drivers already strain the CPU more. As most of the gameworks features, and phsyx, are run on the CPU and don't have GPU acceleration.

 

Look at these results.

 

 

You can clearly see the Nvidia drivers getting almost double the amount of drawcalls out of the DX11 API. Meaning they're much more efficient than AMD's DX11 drivers and have less overhead.

 

The difficulty with this is, GPU testing is done with fast CPU's. Usually 4790K's or 5960X's running overclocks to remove any form of CPU bottleneck. But that also means they have way more capacity to deal with inefficient drivers than say... locked i5's, i3's or AMD FX CPU's. Meaning the R9-380 which, suggested by this forum all the time, does so well vs. the GTX 960...doesn't acually perform that well in the budget build it's being placed in. And that pairing a R9-380 with a 'lowly' i5-4460 might not provide better results as the same i5-4460 + gtx960. I say lowly, because next to the 5960X @ 4.5ghz, the 4460 is kinda lowly

 

But why is the e.g. the 390 getting 2-3 more fps than a 970? And if the benchmarks are more or less identical between similarly priced gpus then the main factor for me at least how stable the drivers are

[Majestic]

But why is the e.g. the 390 getting 2-3 more fps than a 970? And if the benchmarks are more or less identical between similarly priced gpus then the main factor for me at least how stable the drivers are

 

A. you didn't read my post carefully enough. B. most benchmarks don't factor in overclocks, and you can't overclock AMD cards the way you can Maxwell chips.

 

Also, don't quote entire paragraphs. Not only does it add extra clutter, any adjustments made to the post are not taken into account.

[IMPERIUS]

A. you didn't read my post carefully enough. B. most benchmarks don't factor in overclocks, and you can't overclock AMD cards the way you can Maxwell chips.

 

Also, don't quote entire paragraphs. Not only does it add extra clutter, any adjustments made to the post are not taken into account.

 

look at his benches: 

[Majestic]

look at his benches: 

 

Yeah expected that benchmark. Granted, that 1442 mhz is a really, really pisspoor OC for a 970. Not sure how representative that is.

 

And jay is als running a very high-end CPU at overclock. So this enforces my earlier comments.

[D2ultima]

nVidia's drivers have been absolute hell for the last few months, since 350.12 drivers to be precise. A lot of users have had a lot of crashes and TDRs and nVidia has acknowledged and claimed to be trying to fix the issues. I think the latest drivers are mostly problem-free, but I haven't looked at reports for them yet.

 

NORMALLY nVidia's drivers are more frequent and quicker timed to bigger game releases. AMD's drivers, while averaging one a month, are usually betas and the number of WHQL releases are quite little.

 

Right now? nVidia is a landmine and it's impossible if your experience will be good or bad.

AMD however is mostly stable except for some low-util and frametime issues I've heard of with their Fiji cards.

 

Majestic's data also proves how bad AMD is with DX11 in CPU-heavy titles. Right now it's a "pick your poison". Normally nVidia is the better driver maker, but even if their drivers are more efficient and have more features, if you're getting random crashes by having nothing but google chrome open, that doesn't really matter.

[Briggsy]

Haven't had a bad driver since switching to AMD 2 years ago. Only driver crashes I've had for AMD happen while overclocking. That being said, Nvidia drivers recover more gracefully after they crash, but in my experience the crashes were more frequent, and not necessarily due to overclocking.

[Humbug]

Bought an R9 290 just over a year ago. Been the best gaming of my life since then. Rock solid, fast, silky smooth frame delivery in every game I try...

Anyway on topic. On the longterm ideally we want to get away from ever having to even discuss this topic. It's ridiculous that gamers need to change GPU drivers every month in order to optimize for every new AAA game. Vulkan and DX12 will hopefully start moving us away from that. See the below post from a former Nvidia employee.

The first lesson is: Nearly every game ships broken. We're talking major AAA titles from vendors who are everyday names in the industry. In some cases, we're talking about blatant violations of API rules - one D3D9 game never even called BeginFrame/EndFrame. Some are mistakes or oversights - one shipped bad shaders that heavily impacted performance on NV drivers. These things were day to day occurrences that went into a bug tracker. Then somebody would go in, find out what the game screwed up, and patch the driver to deal with it. There are lots of optional patches already in the driver that are simply toggled on or off as per-game settings, and then hacks that are more specific to games - up to and including total replacement of the shipping shaders with custom versions by the driver team. Ever wondered why nearly every major game release is accompanied by a matching driver release from AMD and/or NVIDIA? There you go.

The second lesson: The driver is gigantic. Think 1-2 million lines of code dealing with the hardware abstraction layers, plus another million per API supported. The backing function for Clear in D3D 9 was close to a thousand lines of just logic dealing with how exactly to respond to the command. It'd then call out to the correct function to actually modify the buffer in question. The level of complexity internally is enormous and winding, and even inside the driver code it can be tricky to work out how exactly you get to the fast-path behaviors. Additionally the APIs don't do a great job of matching the hardware, which means that even in the best cases the driver is covering up for a LOT of things you don't know about. There are many, many shadow operations and shadow copies of things down there.

The third lesson: It's unthreadable. The IHVs sat down starting from maybe circa 2005, and built tons of multithreading into the driver internally. They had some of the best kernel/driver engineers in the world to do it, and literally thousands of full blown real world test cases. They squeezed that system dry, and within the existing drivers and APIs it is impossible to get more than trivial gains out of any application side multithreading. If Futuremark can only get 5% in a trivial test case, the rest of us have no chance.

The fourth lesson: Multi GPU (SLI/CrossfireX) is fucking complicated. You cannot begin to conceive of the number of failure cases that are involved until you see them in person. I suspect that more than half of the total software effort within the IHVs is dedicated strictly to making multi-GPU setups work with existing games. (And I don't even know what the hardware side looks like.) If you've ever tried to independently build an app that uses multi GPU - especially if, god help you, you tried to do it in OpenGL - you may have discovered this insane rabbit hole. There is ONE fast path, and it's the narrowest path of all. Take lessons 1 and 2, and magnify them enormously.

Deep breath.

Ultimately, the new APIs are designed to cure all four of these problems.

* Why are games broken? Because the APIs are complex, and validation varies from decent (D3D 11) to poor (D3D 9) to catastrophic (OpenGL). There are lots of ways to hit slow paths without knowing anything has gone awry, and often the driver writers already know what mistakes you're going to make and are dynamically patching in workarounds for the common cases.

* Maintaining the drivers with the current wide surface area is tricky. Although AMD and NV have the resources to do it, the smaller IHVs (Intel, PowerVR, Qualcomm, etc) simply cannot keep up with the necessary investment. More importantly, explaining to devs the correct way to write their render pipelines has become borderline impossible. There's too many failure cases. it's been understood for quite a few years now that you cannot max out the performance of any given GPU without having someone from NVIDIA or AMD physically grab your game source code, load it on a dev driver, and do a hands-on analysis. These are the vanishingly few people who have actually seen the source to a game, the driver it's running on, and the Windows kernel it's running on, and the full specs for the hardware. Nobody else has that kind of access or engineering ability.

* Threading is just a catastrophe and is being rethought from the ground up. This requires a lot of the abstractions to be stripped away or retooled, because the old ones required too much driver intervention to be properly threadable in the first place.

* Multi-GPU is becoming explicit. For the last ten years, it has been AMD and NV's goal to make multi-GPU setups completely transparent to everybody, and it's become clear that for some subset of developers, this is just making our jobs harder. The driver has to apply imperfect heuristics to guess what the game is doing, and the game in turn has to do peculiar things in order to trigger the right heuristics. Again, for the big games somebody sits down and matches the two manually.

Part of the goal is simply to stop hiding what's actually going on in the software from game programmers. Debugging drivers has never been possible for us, which meant a lot of poking and prodding and experimenting to figure out exactly what it is that is making the render pipeline of a game slow. The IHVs certainly weren't willing to disclose these things publicly either, as they were considered critical to competitive advantage. (Sure they are guys. Sure they are.) So the game is guessing what the driver is doing, the driver is guessing what the game is doing, and the whole mess could be avoided if the drivers just wouldn't work so hard trying to protect us.

So why didn't we do this years ago? Well, there are a lot of politics involved (cough Longs Peak) and some hardware aspects but ultimately what it comes down to is the new models are hard to code for. Microsoft and ARB never wanted to subject us to manually compiling shaders against the correct render states, setting the whole thing invariant, configuring heaps and tables, etc. Segfaulting a GPU isn't a fun experience. You can't trap that in a (user space) debugger. So ... the subtext that a lot of people aren't calling out explicitly is that this round of new APIs has been done in cooperation with the big engines. The Mantle spec is effectively written by Johan Andersson at DICE, and the Khronos Vulkan spec basically pulls Aras P at Unity, Niklas S at Epic, and a couple guys at Valve into the fold.

Three out of those four just made their engines public and free with minimal backend financial obligation.

http://www.gamedev.net/topic/666419-what-are-your-opinions-on-dx12vulkanmantle/

[IMPERIUS]

snip

 

yeah I agree, the software must catch up to the hardware

[Prysin]

Hey guys,

I've been hearing a lot of bad stuff about nvidia drivers latelay, the biggest problem being a 960 beating a 780 in the witcher 3 and just general problems all around. But AMD carries this myth of bad drivers his whole life. Personally I haven't had any problems with them.

I'd love to hear your thought on the matter

ive been using AMD for 3+ years now.

To be more precise, ive been using AMD Crossfire (two cards working together) for 3+ years now.

Honestly, i have had only 5 driver related fuckups, with crossfire, since i built my gaming rig.

The whole "AMD DRIVERS SUCKS" is a myth from the olden days, like pre HD 7000 series (which was 2010 i think), that has been kept alive by the Nvidia fanboys out there.

 

Both companies do fuck up their drivers from time to time. Lately, its been Nvidia, unfortunatly, they are on a roll, with over 5 bad drivers in a row now. Even though i am on Team Red, i hope Nvidia get their shit together soon as the way they are going, they are going to switch place with AMD eventually.

 

Even with better performance, if the end users and mass media experiences a lot of issues, your sales will drop like a rock.

[Edgar R. Zakarian]

Hey guys,

I've been hearing a lot of bad stuff about nvidia drivers latelay, the biggest problem being a 960 beating a 780 in the witcher 3 and just general problems all around. But AMD carries this myth of bad drivers his whole life. Personally I haven't had any problems with them.

I'd love to hear your thought on the matter

Never really had any driver issues with AMD, only lack of crossfire support for some of the games.

So using single card, there are no real issues in my experience