SEO
vBulletin Search Engine Optimization
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
01-16-2008, 02:59 PM
| ||||
| ||||
 What's the difference between Elite3D -m3 and -m6? Title pretty much says it all. I've just bought a U60 (the result of answers from my U5 vs U 60 post) and just wondered what the -m6 does that the -m3 does not? I had a -m3 before, but swapped it for a Creator3D, since I wanted 1600x1200. I suspect I'll run headless 99% of the time, so are a bit less concerned now, but just wondered what the differences were. Can't find out from the Framebuffer FAQ. -- Dave K http://www.southminster-branch-line.org.uk/ Please note my email address changes periodically to avoid spam. It is always of the form: month-year@domain. Hitting reply will work for a couple of months only. Later set it manually. The month is always written in 3 letters (e.g. Jan, not January etc)      |
|
01-16-2008, 02:59 PM
| |||
| |||
| Re: What's the difference between Elite3D -m3 and -m6? Dave (from the UK) wrote: > Title pretty much says it all. I've just bought a U60 (the result of > answers from my U5 vs U 60 post) and just wondered what the -m6 does > that the -m3 does not? I had a -m3 before, but swapped it for a > Creator3D, since I wanted 1600x1200. > > I suspect I'll run headless 99% of the time, so are a bit less concerned > now, but just wondered what the differences were. Can't find out from > the Framebuffer FAQ. > The Elite3D-M3 has 3 floating processor units, and the -M6 has 6 floating processor units. The -M6 would be faster for processing 3D vector data, twice the amount of floating processor units would allow for double "Z" buffering, basically while displaying one frame (using 3 floating processor units) and calculating in advance the next frame with the additional 3 floating processor units. Paul |
|
01-16-2008, 02:59 PM
| |||
| |||
| Re: What's the difference between Elite3D -m3 and -m6? HI, Paul Gress wrote: > Dave (from the UK) wrote: > >> Title pretty much says it all. I've just bought a U60 (the result of >> answers from my U5 vs U 60 post) and just wondered what the -m6 does >> that the -m3 does not? I had a -m3 before, but swapped it for a >> Creator3D, since I wanted 1600x1200. >> >> I suspect I'll run headless 99% of the time, so are a bit less >> concerned now, but just wondered what the differences were. Can't find >> out from the Framebuffer FAQ. >> > > > The Elite3D-M3 has 3 floating processor units, and the -M6 has 6 > floating processor units. The -M6 would be faster for processing 3D > vector data, twice the amount of floating processor units would allow > for double "Z" buffering, basically while displaying one frame (using 3 > floating processor units) and calculating in advance the next frame with > the additional 3 floating processor units. > > Paul You wrote that 3 FPU are used during display in double buffering mode, what are they doing during display? Is not "display" just a DMA to the DAC from the refresh memory(read active memory half in double buffering mode) while the FPU and CPU are using the (write active half of the display memory) for creating the next memory buffer? /michael |
|
01-16-2008, 03:00 PM
| |||
| |||
| Re: What's the difference between Elite3D -m3 and -m6? Michael Laajanen wrote: > HI, > > Paul Gress wrote: >> Dave (from the UK) wrote: >> >>> Title pretty much says it all. I've just bought a U60 (the result of >>> answers from my U5 vs U 60 post) and just wondered what the -m6 does >>> that the -m3 does not? I had a -m3 before, but swapped it for a >>> Creator3D, since I wanted 1600x1200. >>> >>> I suspect I'll run headless 99% of the time, so are a bit less >>> concerned now, but just wondered what the differences were. Can't >>> find out from the Framebuffer FAQ. >>> >> >> >> The Elite3D-M3 has 3 floating processor units, and the -M6 has 6 >> floating processor units. The -M6 would be faster for processing 3D >> vector data, twice the amount of floating processor units would allow >> for double "Z" buffering, basically while displaying one frame (using >> 3 floating processor units) and calculating in advance the next frame >> with the additional 3 floating processor units. >> >> Paul > > You wrote that 3 FPU are used during display in double buffering mode, > what are they doing during display? > > Is not "display" just a DMA to the DAC from the refresh memory(read > active memory half in double buffering mode) while the FPU and CPU are > using the (write active half of the display memory) for creating the > next memory buffer? > > > /michael Maybe I got a little ahead of myself. The 6 FPU's are used for calculating the vector rendering. Alternating 3 at a time. So 3 are working what will currently be displayed, and 3 for the future frame. I guess the display would be DMA to the DAC, but this card is a UPA interface and I'm not sure if it's DMA. Paul |
|
01-16-2008, 03:00 PM
| ||||
| ||||
| Re: What's the difference between Elite3D -m3 and -m6? Paul Gress wrote: > Michael Laajanen wrote: > >> HI, >> >> Paul Gress wrote: >> >>> Dave (from the UK) wrote: >>> <snip> >> >> You wrote that 3 FPU are used during display in double buffering mode, >> what are they doing during display? >> >> Is not "display" just a DMA to the DAC from the refresh memory(read >> active memory half in double buffering mode) while the FPU and CPU are >> using the (write active half of the display memory) for creating the >> next memory buffer? >> >> >> /michael > > > Maybe I got a little ahead of myself. The 6 FPU's are used for > calculating the vector rendering. Alternating 3 at a time. So 3 are > working what will currently be displayed, and 3 for the future frame. I > guess the display would be DMA to the DAC, but this card is a UPA > interface and I'm not sure if it's DMA. > > Paul Thanks for answering, But I still not get it it sounds like tripple buffering what you explain? Double buffering is from what i know used when you don't want to have "not completed pictures" in on the screen, so you set a "display mask" to display one memory page while the other is getting updated, once its done you change the "display mask" to the new and start updating the previous page. The "DMA" from the active display page to the Video DAC is done locally on the graphics adapter in this case so no use on the UPA bus is done. /michael |
Linear Mode
