Next-Gen GPU Progress Slowing As It Aims for 20 nm and Beyond 91
JoshMST writes "Why are we in the middle of GPU-renaming hell? AMD may be releasing a new 28-nm Hawaii chip in the next few days, but it is still based on the same 28-nm process that the original HD 7970 debuted on nearly two years ago. Quick and easy (relative terms) process node transitions are probably a thing of the past. 20-nm lines applicable to large ASICs are not being opened until mid-2014. 'AMD and NVIDIA will have to do a lot of work to implement next generation features without breaking transistor budgets. They will have to do more with less, essentially. Either that or we will just have to deal with a much slower introduction of next generation parts.' It's amazing how far the graphics industry has come in the past 18 years, but the challenges ahead are greater than ever."
Intel (Score:3, Informative)
Meanwhile, Intel is about to give us 15 core Ivy-bridge Xeons [wikipedia.org]. A year from now we'll have at least that many Haswell core Xeons, given that they have the same 22nm feature size.
How many cores will 14nm [slashdot.org] Broadwell parts give us (once they sort the yield problems?) You may expect to see 4-5 billion transistor CPUs in the next few years.
Yay for Moore's law.
Geometry, shaders, physics (Score:4, Informative)
Re:I want better 2D performance (Score:5, Informative)
Re:This is good news for me (Score:4, Informative)
Re:Not true (Score:5, Informative)
Not totally true. Stroke/path/fill rasterization work is not supported by current 3D rendering APIs (and thus not accelerated by 3d hardware). Right now the stroke/path/fill rasterization is done on the CPU and merely 2D blit-ed to the frame buffer by the GPU. The CPU could of course attempt convert the stroke/path into triangles and then use the GPU to rasterize those triangles (with some level of efficiency), but that's a far cry from "proper, full-featured 2D".
Fonts are special cased in that glyphs are cached, but small font rasterization isn't generally possible to do with triangle rasterization (because of the glyph hints).
Since SW doesn't even attempt to use HW for modern 2D operations, it will likely be a long time before HW will support this kind of stuff...
A - anything that you can't do by tesselating to triangles could be done with OpenCL or CUDA. You could, for example, assign OpenCL kernels where each instance rasterizes one stroke and composite the results or something similar, and exploit the paralellism of the GPU. But, it would be inconvenient to write. Especially since most PDF viewers don't even bother with effective parallelism in their software rasterizers.
B - you can do anything by tesselating to triangles.
Re:Intel (Score:4, Informative)
Exactly. The new Intel GPUs (those on Haswell) do better than entry level NVidia and ATI dedicated graphics.
Re:Why dribble about GPUs? (Score:5, Informative)
Some nerdy sheeple won't believe what I've just said about Intel's lies. Well Intel gets 10 million transistors per mm2 on its 22nm process, and AMD, via TSMC, gets 14+ million on the larger 28nm process. Defies all concept of maths when Intel CLAIMS a smaller process, but gets far less transistors per area against a larger process.
Comparing apples and oranges are we? Yes, AMD gets 12.2 million/mm^2 on a GPU (7970 is 4.313 billion transistors on 352 mm^2) but CPU transistor density is a lot lower for everybody. The latest Haswell (22nm) has 1.4B transistors in 177mm^2 or about 7.9 million/mm^2, but AMD's Richland (32nm) has only 1.3B transistors in 246mm^2. or 5.3 million/mm^2. Their 28nm CPUs aren't out yet but they'll still have lower transistor density than Intel's 22nm and at this rate they'll be competing against the even smaller Broadwell, though I agree it's probably not true 14nm. Very well formulated post though that appears plausible and posted as AC, paid AMD shill or desperate investor?