4 ms·
>- seems to be based on 2000-era technology Could you elaborate?
by fosap 13y ago
>- seems to be based on 2000-era technology
Could you elaborate?
- agumonkey 13y agoI guess he meant that until the latest goal, it's a fixed rendering pipeline gpu. ps: instead of parallel compute units for vertex/pixel shader. ps2: I'd like to see an open hardware RISC movement for GPU. I wish for a 'simple' parallel array of visual oriented floating point primitives, very small and regular, but very open so that people can write compilers and drivers with ease. As opposed to very potent but obscure gpu followed with crippled drivers. Just enough to get fast yet low power css/dom like systems (with little bonus like path rendering) to help lift the visual work away from RPI-class SoCs.
- sitkack 13y agoWhich struck me as totally backwards. With embedded multipliers available in nearly all levels of FPGAs, I think an OpenCL/HSAIL target would be a more generic solution that could enable many of the hardcoded features in software.
- agumonkey 13y agoHa, you summarized what I was trying to express while I was editing. Completely agree.
- pjc50 13y agoIf you want a large set of general purpose RISC cores, check out Tilera and Netronome. The Raspberry Pi GPU is actually very powerful, but not in ways that help you with CSS/DOM; I don't think there's much work been done to accelerate layout in 3D accelerators other than at the very basic level of glyph rendering and image uncompression.
- agumonkey 13y agoActually the rpi case is fueling my desire. I know that the VideoCore part of the SoC is very potent but it's encapsulated beneath too many layers (especially considering the goal of the rpi, I would love to have simple access to all the gpu).
- lambda 13y agoThe goal is to produce a re-implementation of the #9 Ticket to Ride IV card (https://en.wikipedia.org/wiki/Number_Nine_Visual_Technology https://en.wikipedia.org/wiki/Number_Nine_Visual_Technology) which is a fixed-function graphics card released in 1998. As they say in the first paragraph, "The reason behind this was to provide a binary compatible graphics core for vertical markets: Medical Imaging, Military, Industrial, and Server products." This is not something that will be remotely competitive with modern GPUs. Their $1,000,000 stretch goal would be to implement a modern shader engine, but given that they aren't offering any actual hardware, I'm not sure how they hope to get to that point.
- agumonkey 13y agoI missed the 'history' paragraph. I remember the glorious #9 days. But this is so old I'm thinking it would be easier to have virtual driver translation. Even with overhead a phone gpu could outperform easily.
- pjc50 13y agoYou've got it. The $200k level gives you a 2D graphics card from 1998. It looks like their original target market was people who want an embedded emulation of this sort of ancient system so they can run DOS, Windows 95, or ancient Linux/X11 on it. Markets in which changing the software is horribly expensive due to compliance costs. The $400k goal targets Direct3D 8. That appears to date to about 2000: http://www.gamedev.net/page/resources/_/technical/directx-and-xna/directx-8-graphics-and-video-a-fresh-start-r1247 http://www.gamedev.net/page/resources/_/technical/directx-an... (I'm impressed that article is still up from 2000!) $1m is not totally unreasonable for the shader-based design at commercial rates. Last time I did a back-of-the-envelope calculation for a shader-capable fully open graphics card delivering working silicon, I came up with two years, $2.5m, and a hiring shortlist.
- cwt137 13y agoIf you look at the demo video at the end, you can see that they are using WordPerfect on Win 98 or Win 2k. I thought this was some kind of joke. But apparently not. So, I believe it when fosap says 2000-era technology.