When I develop shader code on my machine I often find myself in the situation where the shader works perfectly on my machine, but on other graphic cards, drivers, operating systems, etc. it doesn't.
How to achieve compatibility of shaders?
I see a few approches:
Test on many different systems. But which systems to choose? Testing with every card on every OS and every driver is not realistic. Maybe we can assume that the vendors care for backward compatibility? In this case testing with old cards and drivers might be sufficient.
Ask the driver for a specific version and core profile. This helps a bit, but the drivers seem very lenient, allowing me to code things that aren't in the spec.
Code checkers that check the code for strict compatibility with a certain spec. There don't seen to be such tools around.
Don't bother and wait for bug-reports from users.Yet the error messages generated by the drivers are rather poor, and the observed behaviour might be as un-insightful as a black screen.
I'm targeting Win/Linux/OSX platforms. Not consoles.
Khronos has released OpenGL / OpenGL ES Reference Compiler that can be used to validate shader's source code. From the site:
The primary purpose of the reference compiler is to identify shader
portability issues.
Related
OpenGL Stenciling, seperating ref from value written?
In the answer to this question, a vender specific extension GL_REPLACE_VALUE_AMD is able to do exactly what I'm struggling to do in OpenGL, but I'm worried it will limit what computers and platforms I want my program to run on, and I've had no luck researching where it would not be available.
My goal is for the program to run on any computer that supports OpenGL 2.0, without any functional differences between them. Should I compile a program that uses this extension, what computers/platforms in this set would no longer be able to run the program without problems, if any?
The fact that it's a vendor extension should be an immediate clue that there's a good chance that you'd be limiting yourself to that vendor's hardware. It's not a 100% guarantee; NV_texture_barrier has been implemented for years on pretty much anything that can run GL 3.3 or better.
Further research indicates that the publication date for that extension is from 2012. That suggests that the extension would likely be implemented by more recent, GL 4.x-capable hardware.
If you want more accurate information, there are databases of extension usage that give a clearer picture. From this, we see that the extension is only implemented on AMD hardware. While it is available on AMD's GL 3.x-class hardware, it is not available on any of AMD's 2.x class hardware.
So if your goal is to support GL 2.0 (why not 2.1?) as a maximum, then you can't use that extension.
As I understand GPU vendors defined standard interface to be used by OS Developers to communicate with their specific driver. So DirectX and OpenGL are just wrappers for that interface. When OS developers decide to create new version of Graphic API , GPU vendors expand their interface (new routines are faster and older ones are left for compatibility issues) and OS developers use this new part of interface.
So, when it is said that GPU vendors' support for DirectX is better than for OpenGL, does it simply mean that GPU vendors primarily take into account Microsoft's future plans of developing DirectX API structure and adjust future development of this interface to their needs? Or there is some technical reasons before this?
As I understand GPU vendors defined standard interface to be used by OS Developers to communicate with their specific driver. So DirectX and OpenGL are just wrappers for that interface.
No, not really. DirectX and OpenGL are just specifications that define APIs. But a specification is nothing more than a document, not software. The OpenGL API specification is controlled by Khronos, the DirectX API specification is controlled by Microsoft. Each OS then defines a so called ABI (Application Binary Interface) that specifies which system level APIs are supported by the OS (OpenGL and DirectX are system level APIs) and what rules an actual implementation must adhere to, when being run on the OS in question.
The actual OpenGL or Direct3D implementation happens in the hardware's drivers (and in fact the hardware itself is part of the implementation as well).
When OS developers decide to create new version of Graphic API , GPU vendors expand their interface
In fact it's the other way round: Most of the graphic APIs specifications are laid out by the graphics hardware vendors. After all they are close to where the rubber hits the road. In the case of Khronos the GPU makers are part of the controlling group of Khronos. In the case of DirectX the hardware makers submit drafts to and review the changes and suggestions made by Microsoft. But in the end each new APIs release reflects the common denominator of the capabilities of the next hardware generation in development.
So, when it is said that GPU vendors' support for DirectX is better than for OpenGL, does it simply mean that GPU vendors primarily take into account Microsoft's future plans of developing DirectX API structure and adjust future development of this interface to their needs?
No, it means that each GPU vendor implements his own version of OpenGL and the Direct3D backend, which is where all the magic happens. However OpenGL puts a lot of emphasis on backward compatibility and ease of transition to newer functionality. Direct3D development OTOH is quick in cutting the ties with earlier versions. This also means that full blown compatibility profile OpenGL implementations are quite complex beasts. That's also the reason why recent versions of OpenGL core profiles did (overdue) work in cutting down support for legacy features; this reduction of API complexity is also quite a liberating thing for developers. If you develop purely for a core profile it simplifies a lot of things; for example you no longer have to worry about a plethora of internal state when writing plugin.
Another factor is, that for Direct3D there's exactly one shader compiler, which is not part of the driver infrastructure / implementation itself, but gets run at program build time. OpenGL implementations however must implement their own GLSL shader compiler, which complicates things. IMHO the lack of a unified AST or immediate shader code is one of the major shortcomings of OpenGL.
There is not a 1:1 correspondence between the graphics hardware abstraction and graphics API like OpenGL and Direct3D. WDDM, which is Windows Vista's driver model defines things like common scheduling, memory management, etc. so that DirectX and OpenGL applications work interoperably, but very little of the design of DirectX, OpenGL or GPUs in general has to do with this. Think of it like the kernel, nobody creates a CPU specifically to run it, and you do not have to re-compile the kernel everytime a new iteration of a processor architecture comes out that adds a new subset of instructions.
Application developers and IHVs (GPU vendors, as you call them) are the ones who primarily deal with changes to GPU architecture. It may appear that the operating system has more to do with the equation than it actually does because Microsoft (more so) and Apple--who both maintain their own proprietary operating systems--are influential in the design of DirectX and OpenGL. These days OpenGL closely follows the development of commodity desktop GPU hardware, but this was not always the case - it contains baggage from the days of custom SGI workstations and lots of things in compatibility profiles have not been hardware native on desktop GPUs in decades. DirectX, on the other hand, has always followed desktop hardware. It used to be if you wanted an indication of where desktop GPUs were headed, D3D was a good marker.
OpenGL is arguably more complicated than DirectX because until recently it never let go of anything, whereas DirectX radically redefined the API and stripped legacy support with every iteration. Both APIs have settled down in recent years, but D3D still maintains a bit of an edge considering it only has to be implemented on a single platform and Microsoft writes the one and only shader compiler. If anything, the shader compiler and minimal feature set (void of legacy baggage) in D3D is probably why you get the impression that vendors support it better.
With the emergence of AMD Mantle, the desktop picture might change again (think back to the days of 3Dfx and Glide)... it certainly goes to show that OS developers have very little to do with graphics API design. NV and AMD both have proprietary APIs on the PS3, GameCube/Wii/WiiU, and PS4 that they have to implement in addition to D3D and OpenGL on the desktop, so the overall picture is much broader than you think.
While AMD is following the OpenGL specification very strict, nVidia often works even when the specification is not followed. One example is that nVidia supports element incides (used in glDrawElements) on the CPU memory, whereas AMD only supports element indices from a element array buffer.
My question is: Is there a way to enforce strict OpenGL behaviour using a nVidia driver? Currently I'm interested in a solution for a Windows/OpenGL 3.2/FreeGlut/GLEW setup.
Edit: If it is not possible to enforce strict behaviour on the driver itself - is there some OpenGL proxy that guarantees strict behaviour (such as GLIntercept)
No vendor enforces the specification strictly. Be it AMD, nVidia, Intel, PowerVR, ... they all have their idiosyncrasies and you have to learn to live with them, sadly. That is one of the annoying things about having each vendor implement their own GLSL compiler, as opposed to Microsoft implementing the one and only HLSL compiler in D3D.
The ANGLE project tries to mitigate this to a certain extent by providing a single shader validator shared across many of the major web browsers, but it is an uphill battle and this only applies to WebGL for the most part. You will always have implementation differences when every vendor implements the entire API themselves.
Now that Khronos group has seriously taken on the task of establishing a set of conformance tests for desktop OpenGL like they have for WebGL / OpenGL ES, things might start to get a little bit better. But forcing a driver to operate in a strict conformance mode is not really a standard thing - there may be #pragmas and such that hint the compiler to behave more strictly, but these are all vendor specific.
By the way, I realize this question has nothing to do with GLSL per-se, but it was the best example I could give.
Unfortunately, the only way you can be sure that your OpenGL code will work on your target hardware is to test it. In theory simply writing standard compliant code should work everywhere, but sadly this isn't always the case.
I am having problems getting my GLSL shaders to work on both AMD and Nvidia hardware.
I am not looking for help fixing a particular shader, but how to generally avoid getting these problems. Is it possible to check if a shader will compile on AMD/Nvidia drivers without running the application on a machine with the respective hardware and actually trying it?
I know, in the end, testing is the only way to be sure, but during development I would like to at least avoid the obvious problems.
Everyone using GLSL must have these problems, so why can't I find a good way to fix them?
Is it possible to check if a shader will compile on AMD/Nvidia drivers without running the application on a machine with the respective hardware and actually trying it?
No. If you are going to be serious about developing applications, testing on a variety of hardware is the only reliable way to go about it. And if you're not going to be serious, then who cares.
Generally speaking, the easiest way to handle this for a small team is to avoid the problem altogether. Most driver incompatibilities come from attempting to do something unorthodox: passing arrays as output/input varying variables, passing matrices as attributes, using more recent driver features, etc. So... don't do that. Use only solid, safe stuff that's been around in GLSL and has been almost certainly used in real-world OpenGL applications.
using NVidias's NVEmulate and AMD's GPU ShaderAnalyzer could be an option.
AMD's GPU ShaderAnalyzer is a stand-alone GLSL/HLSL compiler. NVidias's NVEmulate is a tool to emulate the features of different (better) NVidia graphic cards in software. So if you have an NVidia card, you can simply run your program to test it (possibly emulating another NVidia card with the NVEmulate) and use the ShaderAnalyser to see if your shader compile on AMD cards.
If your shader runs on AMD, it will most probably run on NVidia. You can still test this with cgc (NVidias stand-alone Cg compiller, part of the Cg Toolkit) it compiles GLSL and Cg code to binary or cross-compile it to HLSL. This is the compiler which NVidia drivers use for GLSL anyway.
The bonus is that you can also see the binary/assembly code of your shader, which is very helpful for low-level optimizations.
What no tool can tell you is, if the shader works (as expected) on different hardware.. I recently found out that some new AMD driver don't handle default uniform values properly... without a warning or error message... but this is a different story.
So at some point you have to test your code on the target hardware.
I have developed a program which makes use of many of OpenGL's aspects - ranging from both rather new to deprecated functionalities, and want to ensure that it works correctly on the great majority of machines - especially on ones with outdated graphics cards.
What is the best way to maximize the (backwards)compatibility of an OpenGL application?
How can I test my program for compatibility with older hardware without actually having a test machine with older hardware?
What ways are there to find the underlying causes of the issues which may be encountered during compatibility testing?
What is the best way to maximize the (backwards)compatibility of an OpenGL application?
Define "compatibility"? If you want an application to run on as much hardware as possible, then you basically have to give up on shaders entirely and stick to about GL 1.4. The main confounding issue here are Intel driver bugs; many pieces of older Intel hardware will claim support for GL 2.0 or 2.1, but they have innumerable failings in this support.
How can I test my program for compatibility with older hardware without actually having a test machine with older hardware?
You don't. Compatibility with old hardware is about more than just sticking to a standard. It's about making sure that your program doesn't encounter driver bugs. And the only way to do that is to actually test on the hardware of interest.
What ways are there to find the underlying causes of the issues which may be encountered during compatibility testing?
Test the same code on recent hardware. If it has the same failures, then the problem is likely in your code. If it works fine on recent hardware but fails on older stuff, then the problem is almost certainly a driver bug with old hardware drivers.
Develop a workaround.
Well, the best way to maximize the backwards compatibility and to get a powerful tool on tracking down target machine's functionality (imho) is to use something like GLEW: The OpenGL Extension Wrangler Library. It will load OpenGL version-specific functions for you and you can test if they are supported by user's system (or, more correctly, by video drivers).
This library is very simple in use, it is well documented and you can google a lot of examples.
So if target machine doesn't have some new opengl functions, you load module named "opengl_old.cpp" (for example), or if it don't have some functionality which is already deprecated (like glBegin(), glEnd()), you'd better go on with "opengl_new.cpp".
Basically the most changes are done in OpenGL 3.0 (and furthermore 3.3) with shaders introduced as the only non-deprecated graphics pipeline, so you can make two opengl modules in your program: one for OpenGL 1&2 and one for OpenGL 3&4. At least I solved this problem in this way in my own code.
To test some functionality you can specify concrete version of OpenGL API to be loaded, when creating context.