News for hackers. Stuff that matters.
Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by shmerl on Thursday April 26, 2018 @10:57AM (#16932430)
AMD should put a good effort into GPUs too, and replace GCN with brand new and efficient architecture.
Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by hajile on Thursday April 26, 2018 @06:58PM (#16936746)
Most of Nvidia's architectural changes have moved it closer to GCN (eg, adding async compute in Pascal). In other areas like multiple compute engines (A reason GCN generally has better VR latency numbers), Nvidia still lags behind (not sure if Volta changes that). AMD generally does quite well on compute tasks compared to Nvidia and is still generally much faster for anything using Vulkan.

http://ext3h.makegames.de/DX12_Compute.html

The big issue with AMD is software and money. Even with the most talented engineers it takes loads of man hours to build up an ecosystem. AMD's is currently valued at 8.3B. Nvidia currently has 7.1B cash on hand. With that kind of money, Nvidia will do things like loan AAA gamemakers some devs to re-write large parts of the game to optimize them for Nvidia graphics cards or build up the CUDA ecosystem.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by ahartmetz on Friday April 27, 2018 @06:41AM (#16939733)
It seems like AMD has hardware bottlenecks (or missing optimizations relative to Nvidia) in the fixed-function geometry processing parts of GCN. The compute benchmarks are always better than the graphics benchmarks. If it was only optimizations in games, there would be outliers in less popular games (Doom is a bit of an outlier but popular). And in any case, Xbox and PS4 consoles use AMD GPUs so game devs have reason to optimize for AMD.

Also, AMD has been improving its Windows drivers in the last 2-3 years according to reports. I use AMD GPUs on Linux where the driver codebase is different - the improvements there are huge. So much so that it may make sense for AMD to switch to that stack on Windows in the future.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by IntelMiner on Thursday April 26, 2018 @01:05PM (#16933663)
ATi hasn't really competed with Nvidia at the top end of the market for a very long time

They've basically thrown in the towel as far as the datacenter went with CUDA

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by shmerl on Thursday April 26, 2018 @01:12PM (#16933699)
So they should make an effort to compete more.
Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by dragontamer on Thursday April 26, 2018 @02:30PM (#16934322)
They're behind in a number of areas.

Their ROCm environment finally is catching up with NVidia's software / CUDA environment. OpenCL was simply a disaster and hampered AMD's role as a serious software solution for years.

I think its more important to get ROCm fully functional with Tensorflow and other technologies (their current path). AMD can compete on price/performance, selling Vega HBM2 chips at ~$1000 to $2000 (compared to NVidia's chips at $8000 for their HBM2-based V100).

With AMD's drivers properly in the Linux kernel and a nicer licensing model, AMD can achieve a position in the server world. But only if developers start coding in ROCm instead of CUDA.

Honestly, I bet you that software is the bigger issue in the near term. No one wants to code in OpenCL / separated C-environment. ROCm achieves CUDA-parity to some degree with "single source" C++ programming and is beginning to become compatible with Tensorflow.

Sure, they'll be slower than NVidia. But at least your python machine learning code will actually run at all. And paying 1/8th the price for ROCm acceleration will be fine as long as AMD is 1/4th the performance or better. That's how you can actually build a value argument.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by se6 on Thursday April 26, 2018 @03:39PM (#16934935)
OpenCL is not a disaster at all. It is just that NVidia were (and still are) too scared to have people move away from their proprietary solutions, so they tried to hide OpenCL has much as they could and only pushed Cuda.

Even today OpenCL is a viable solution for GPU. It works fine on both AMD and NVidia GPUs. It is also pushed a lot by Intel for FPGA, which probably scare even more NVidia.

OpenCL kernels are compiled at runtime, which is brilliant since you can change the kernels code at run time, use constants in the code at the last moment, unroll, etc.. which can gives better performances. (Nvidia only introduced the possibility of having runtime compilation as a preview in Cuda 7!)

The "single source" argument is completely overrated. Furthermore, you can have single source in OpenCL putting the code in strings.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by aseipp on Thursday April 26, 2018 @04:15PM (#16935258)
> It is also pushed a lot by Intel for FPGA, which probably scare even more NVidia.

These tools aren't available for the wide majority of developers, and are still exceptionally difficult to use and maintain without hardware engineers. I'm going to assume you haven't used FPGAs at all? The ones that can compete at the same tasks for GPUs are not as easily available in terms of price, volume, or even over-the-counter availability (be prepared to ask for a lot of quotes), and the tools have only become more accessible very recently -- such as Intel slashing the FPGA OpenCL licensing costs, and Dell EMC shipping them in pre-configured rack units.

> Nvidia only introduced the possibility of having runtime compilation as a preview in Cuda 7

In the mean time, Nvidia also completely dominated the market by actually producing many working middleware libraries and integrations, a solid and working programming model, and continuously refining and delivering on core technology and GPU upgrades. Maybe those things matter more than runtime compilation and speculative claims about peak performance...

> The "single source" argument is completely overrated.

Even new Khronos standards like SYCL (built on OpenCL, and which does look promising, and I'm hoping AMD delivers a toolchain after they get MIOpen more fleshed out) are moving to the single-source model. It's not even that much better, really, but development friction and cost of entry matters more than anything, and Nvidia understood this from day one. They understood it with GameWorks, as well. They plant specialist engineers "in the field" to accelerate the development and adoption of their tech, and they're very good at it.

This is because their core focus is hardware and selling hardware; it's thus in their interest to release "free" tools that require low-effort to buy into, do as much dirty integration work as possible, and basically give people free engineering power -- because it drives their hardware sales. They basically subsidize their software stack in order to drive GPUs.

> Furthermore, you can have single source in OpenCL putting the code in strings.

This is a joke argument, right?

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by dragontamer on Thursday April 26, 2018 @04:31PM (#16935427)
> OpenCL is not a disaster at all.

I'll probably need to be more specific. OpenCL 1.0 through 1.2 is fine, but fell hopelessly behind NVidia's CUDA efforts. NVidia CUDA has more features that lead to proven performance enhancements.

OpenCL 2.0 was the "counterpunch" to bring OpenCL up to CUDA-level features. However, OpenCL 2.0 is virtually stillborn. Only Intel and AMD platforms support OpenCL2.0. Intel Xeon Phi are relatively niche (and their primary advantage seems to be x86 code compatibility anyway. So I doubt you'd be running OpenCL on them).

AMD OpenCL 2.0 support exists, but is rather poor. The OpenCL 2.0 debugger simply is non-functional and you're forced to use lol printfs.

That leaves OpenCL 1.2. Its okay, but it is years behind what modern hardware can do. Its atomic + barrier model is strange compared to proper C++11 Atomics, its missing important features like device-side queuing, shared virtual memory, unified address space (no more copy/paste code just to go from "local" to "private" memory), among other very useful features.

> Even today OpenCL is a viable solution for GPU

OpenCL 1.2 is a viable solution. An old, crusty, and quirky solution, but viable nonetheless. OpenCL 2.0+ is basically dead. And I think only Intel Xeon Phi supports the latest OpenCL 2.2.

I bet you there are more Vulkan compute shaders out there than there are OpenCL 2.0. Indeed, there are rumors that the Khronos project is going to be focusing on Vulkan compute shaders in the future.

> The "single source" argument is completely overrated. Furthermore, you can have single source in OpenCL putting the code in strings.

I like my compile-time errors to be during compile-time. Not during run-time on my client's system. Compiler-bugs in AMD drivers are fixed through device driver updates (!!!) which makes practical deployment of plain-text OpenCL source code far more of a hassle in practice.

Consider this horror story: a compiler bug in some AMD Device Driver versions which cause a segfault on some hardware versions. This is not theoretical: https://community.amd.com/thread/160362.

In practice, deploying OpenCL 1.2 code requires you to test all of the device drivers your client base is reasonably expected to run.

-----

But that's not the only issue.

"Single Source" means that you can define a singular structure in a singular .h file and actually have it guaranteed to work between CPU-code and GPU-code. Data-sharing code is grossly simplified and is perfectly matched.

The C++ AMP model (which has been adopted into AMD's ROCm platform) is grossly superior. You specify a few templates and bam, your source code automatically turns into CPU code OR GPU-code. Extremely useful when sharing routines between the CPU and GPU (like data-packing or unpacking from the buffers)

With that said, AMD clearly cares about OpenCL and the ROCm platform looks like it strongly supports OpenCL through then near term, especially OpenCL 1.2 which seems to have a big codebase.

However, if I were to do any project these days, I'd do it in ROCm's HCC / single-source C++ system or CUDA. OpenCL 1.2 is useful for high-compatibility but has major issues as an environment.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by se6 on Thursday April 26, 2018 @04:50PM (#16935619)
The point I really wanted to make here is that OpenCL is only a disaster because NVidia was scared of the competition it would bring from AMD.
Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by dragontamer on Thursday April 26, 2018 @04:51PM (#16935640)
I'm sure NVidia deserves some blame.

But AMD drivers which cause OpenCL compiler-segfaults and/or infinite loops is a problem that rests squarely on AMD's shoulders.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by se6 on Thursday April 26, 2018 @05:11PM (#16935832)
I have extensively used OpenCL on both AMD and NVidia for a few years and never had such problems. If anything, found a few more bugs with NVidia.
Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by dragontamer on Thursday April 26, 2018 @05:23PM (#16935944)
Interesting. I'll take your anecdote for what its worth.

My personal use case with OpenCL didn't seem to be going very well. I was testing on my personal Rx 290x. While I didn't have the crashing / infinite loop bugs (See LuxRender's "Dave" for details: http://www.luxrender.net/forum/viewtopic.php?f=34&t=11009) that other people had, my #1 issue was with the quality of AMD's OpenCL compiler.

In particular, the -O2 flag would literally break my code. I was doing some bit-level operations, and those bit-level operations were just wrong under -O2. While the -O0 flag was so unoptimized that my code was regularly swapping registers into / out of global memory. At which point the CPU was faster at executing and there was no point in using OpenCL / GPU compute.

It seems like AMD's OpenCL implementation assumes that the kernels would be very small and compact. And it seems to be better designed for floating-point ops. Other programmers online have also complained about AMD's bit-level operations returning erronious results under -O2. My opinion of its compiler was... rather poor... based on my limited exposure. And further research seems to indicate that I wasn't the only one having these issues.

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by se6 on Friday April 27, 2018 @03:13AM (#16938980)
Only did and do floating point for image processing. In fact, looking into my logs, I registered 5 bugs with NVidia in the last 2 years, none with AMD.
Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by IntelMiner on Thursday April 26, 2018 @01:21PM (#16933775)
That's an incredibly ignorant comment

Just because they can trade blows with Intel in the CPU space, doesn't automatically translate to them having equal engineering talent to take on Nvidia on GPU's

Re:AMD's Stock Price Jumps on News of Earnings Spurred by Ryzen by shmerl on Thursday April 26, 2018 @01:28PM (#16933840)
I didn't say they have talent (which they might have anyway). I said they should compete, which means they should find needed talent obviously if they don't have it.