Guest Tony Sperling Posted March 6, 2008 Posted March 6, 2008 Anybody know - does an application that is not SLI aware recognise the combined Video Memory installed on the two cards? Tony. . .
Guest Theo Posted March 6, 2008 Posted March 6, 2008 Re: OT. . . I don't know, but everything I have read on it leads me to believe it is primarily the hardware and the video driver. With 2 video cards one handles the upper part of the screen and the other handles the lower part so the display is essentially twice as fast. I get the impression the applications essentially just see the video processing as it always has; not doing anything special. Tony Sperling wrote: > Anybody know - does an application that is not SLI aware recognise the > combined Video Memory installed on the two cards? > > > Tony. . . > > >
Guest Carlos Posted March 6, 2008 Posted March 6, 2008 RE: OT. . . Tony, My best guess is that apps don't need to be SLI aware. They will just work, no matter how many video cards are in SLI. I think that the video card driver handles everything. FWIW. I bought one XFX8800GT Alpha Dog Edition video card. This is a factory super-overclocked card with the newest NVidia G92 chip. (8800GTX uses an older, super-power consumption chip). Even with that there is no way to run FSX with all the gizmos enabled. If you intend to run FSX at full speed, check before what SLI combo will do it. Carlos "Tony Sperling" wrote: > Anybody know - does an application that is not SLI aware recognise the > combined Video Memory installed on the two cards? > > > Tony. . . > > > >
Guest Tony Sperling Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Yes, I think you are right. The app will be talking to the driver and the driver is seeing one device? But I also do believe that an application can be written to set up it's own SLI behavior and, thus, override the driver. Thanks, Theo! Tony. . . "Theo" <theo@discussions.microsoft.com> wrote in message news:O3yIfY9fIHA.5624@TK2MSFTNGP02.phx.gbl... > I don't know, but everything I have read on it leads me to > believe it is primarily the hardware and the video driver. > With 2 video cards one handles the upper part of the screen > and the other handles the lower part so the display is > essentially twice as fast. I get the impression the > applications essentially just see the video processing as it > always has; not doing anything special. > > > > Tony Sperling wrote: > > Anybody know - does an application that is not SLI aware recognise the > > combined Video Memory installed on the two cards? > > > > > > Tony. . . > > > > > >
Guest Tony Sperling Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Alpha Dog - indeed! :0) I've never heard of that one. No wonder your temp's were rising, with such power! Actually, I've skipped investing in FSX - the Demo ran very well here - but I decided it was FS2000 all over - too much new technology in one go. The next version WILL be first class! (Just like FS 2002). They are already working on it! Tony. . . "Carlos" <Carlos@discussions.microsoft.com> wrote in message news:2DFB1D9E-639F-4E9A-8ED9-FBFAB20FE321@microsoft.com... > Tony, > My best guess is that apps don't need to be SLI aware. > They will just work, no matter how many video cards are in SLI. > I think that the video card driver handles everything. > FWIW. > I bought one XFX8800GT Alpha Dog Edition video card. > This is a factory super-overclocked card with the newest NVidia G92 chip. > (8800GTX uses an older, super-power consumption chip). > Even with that there is no way to run FSX with all the gizmos enabled. > If you intend to run FSX at full speed, check before what SLI combo will do > it. > > Carlos > > "Tony Sperling" wrote: > > > Anybody know - does an application that is not SLI aware recognise the > > combined Video Memory installed on the two cards? > > > > > > Tony. . . > > > > > > > >
Guest jabloomf1230 Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . No. Each card uses its own memory. A program that is not SLI aware will just ignore the 2nd card altogether. Jay Tony Sperling wrote: > Anybody know - does an application that is not SLI aware recognise the > combined Video Memory installed on the two cards? > > > Tony. . . > > >
Guest John Barnes Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Not correct. nVidia whitepaper specifically states 'SLI is completely implemented in NVIDIA's hardware and software, rendering the technology transparent to affected applications, such as games.' Additionally states 'The graphics driver is effectively able to analyze 3D scenes and divide the load sent to each video card according to the fill rate, geometry, and shading demands with relatively little host processor overhead. In some cases, you'll see the screen divided horizontally into halves, but because the whole process is dynamic, workloads constantly shift to reflect varying loads placed on each card. Thus, although one frame might be cut 50/50, if the next is imbalanced, NVIDIA's driver will make adjustments to compensate.' "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > No. Each card uses its own memory. A program that is not SLI aware will > just ignore the 2nd card altogether. > > Jay > > Tony Sperling wrote: >> Anybody know - does an application that is not SLI aware recognise the >> combined Video Memory installed on the two cards? >> >> >> Tony. . . >> >>
Guest Tony Sperling Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Thanks, John! Much as I thought. I could be wasting 'some' cash now, but there is also going to be 'some' instant reward. Let's see, then, if the future brings more SLI or more multi-core GPU's? Tony. . . "John Barnes" <jbarnes@email.net> wrote in message news:OBfdPi$fIHA.4744@TK2MSFTNGP06.phx.gbl... > Not correct. nVidia whitepaper specifically states 'SLI is completely > implemented in NVIDIA's hardware and software, rendering the technology > transparent to affected applications, such as games.' > > Additionally states 'The graphics driver is effectively able to analyze 3D > scenes and divide the load sent to each video card according to the fill > rate, geometry, and shading demands with relatively little host processor > overhead. In some cases, you'll see the screen divided horizontally into > halves, but because the whole process is dynamic, workloads constantly shift > to reflect varying loads placed on each card. Thus, although one frame might > be cut 50/50, if the next is imbalanced, NVIDIA's driver will make > adjustments to compensate.' > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > > No. Each card uses its own memory. A program that is not SLI aware will > > just ignore the 2nd card altogether. > > > > Jay > > > > Tony Sperling wrote: > >> Anybody know - does an application that is not SLI aware recognise the > >> combined Video Memory installed on the two cards? > >> > >> > >> Tony. . . > >> > >>
Guest Carlos Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Tony, The upcoming NVidia 9800 GX2 (dual-core GPU) might just be what you need. You will certainly need to shell out a lotta cash as opposed to "some" cash, though! :) Carlos "Tony Sperling" wrote: > Thanks, John! Much as I thought. > > I could be wasting 'some' cash now, but there is also going to be 'some' > instant reward. Let's see, then, if the future brings more SLI or more > multi-core GPU's? > > > Tony. . . > > > > "John Barnes" <jbarnes@email.net> wrote in message > news:OBfdPi$fIHA.4744@TK2MSFTNGP06.phx.gbl... > > Not correct. nVidia whitepaper specifically states 'SLI is completely > > implemented in NVIDIA's hardware and software, rendering the technology > > transparent to affected applications, such as games.' > > > > Additionally states 'The graphics driver is effectively able to analyze 3D > > scenes and divide the load sent to each video card according to the fill > > rate, geometry, and shading demands with relatively little host processor > > overhead. In some cases, you'll see the screen divided horizontally into > > halves, but because the whole process is dynamic, workloads constantly > shift > > to reflect varying loads placed on each card. Thus, although one frame > might > > be cut 50/50, if the next is imbalanced, NVIDIA's driver will make > > adjustments to compensate.' > > > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > > news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > > > No. Each card uses its own memory. A program that is not SLI aware will > > > just ignore the 2nd card altogether. > > > > > > Jay > > > > > > Tony Sperling wrote: > > >> Anybody know - does an application that is not SLI aware recognise the > > >> combined Video Memory installed on the two cards? > > >> > > >> > > >> Tony. . . > > >> > > >> > > >
Guest Tony Sperling Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Ha! No doubt. That IS the eternal direction 'evolution' has staked out for us, mere mortals. Honestly, though, I don't know exactly what it is that I am looking for, I was thinking, in this day-and-age, that perhaps Charlie's idea of having an extra of almost everything would warrant an SLI/CrossFire?) solution. Specially since a cheapo extra would be a total waste until you actually needed it. An SLI will be earning it's investment while in use, AND having the beneficial ability of a dismantelment factor in an emergency, to somewhat gloss over the smart-ass factor that inevatibly comes with it as well? It's an interesting idea, though. The '9er' series should have the power we want and a more finely tuned BIOS to fit the newly emerging technologies. I'll go and see what Tom has to say about it (THG!). If they haven't spoken allready, then the word must be practically on the threshold by now. Who knows - in the final end collecting this kind of info could actually save you a bit of money? Tony. . . "Carlos" <Carlos@discussions.microsoft.com> wrote in message news:159B5222-5C84-4861-8813-785F6C065B13@microsoft.com... > Tony, > The upcoming NVidia 9800 GX2 (dual-core GPU) might just be what you need. > You will certainly need to shell out a lotta cash as opposed to "some" cash, > though! > :) > Carlos > > "Tony Sperling" wrote: > > > Thanks, John! Much as I thought. > > > > I could be wasting 'some' cash now, but there is also going to be 'some' > > instant reward. Let's see, then, if the future brings more SLI or more > > multi-core GPU's? > > > > > > Tony. . . > > > > > > > > "John Barnes" <jbarnes@email.net> wrote in message > > news:OBfdPi$fIHA.4744@TK2MSFTNGP06.phx.gbl... > > > Not correct. nVidia whitepaper specifically states 'SLI is completely > > > implemented in NVIDIA's hardware and software, rendering the technology > > > transparent to affected applications, such as games.' > > > > > > Additionally states 'The graphics driver is effectively able to analyze 3D > > > scenes and divide the load sent to each video card according to the fill > > > rate, geometry, and shading demands with relatively little host processor > > > overhead. In some cases, you'll see the screen divided horizontally into > > > halves, but because the whole process is dynamic, workloads constantly > > shift > > > to reflect varying loads placed on each card. Thus, although one frame > > might > > > be cut 50/50, if the next is imbalanced, NVIDIA's driver will make > > > adjustments to compensate.' > > > > > > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > > > news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > > > > No. Each card uses its own memory. A program that is not SLI aware will > > > > just ignore the 2nd card altogether. > > > > > > > > Jay > > > > > > > > Tony Sperling wrote: > > > >> Anybody know - does an application that is not SLI aware recognise the > > > >> combined Video Memory installed on the two cards? > > > >> > > > >> > > > >> Tony. . . > > > >> > > > >> > > > > > >
Guest Carlos Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Tony, The '9er' is a confusing mixed bag of products. 9600GT, with the G94 chip, is a crippled version of the G92 chip (the one that powers my XFX8800GT). Expect -15% of performance. 9500GT, with the G96 chip, is a super crippled version of the G92 chip. You don't want this card. 9800Gx2, with two G92 chips, the dual-core GPU one. You surely want this one. 9800GTX, with the newer G100 chip, the only one that supports DirectX10.1. Its Fill Rate is 99840 MTexels/s, memory bus width is 512bit and memory bandwith is 128 GB/s. Compare this to my 8800GT with 33600 MTexels/s, 256bit and 57.60GB/sec and you will notice that maybe, and I mean maybe, that is the card for FSX!!! :) All the others cards support only DirectX10, without the ".1" at the end. "Tony Sperling" wrote: > Ha! > > No doubt. That IS the eternal direction 'evolution' has staked out for us, > mere mortals. Honestly, though, I don't know exactly what it is that I am > looking for, I was thinking, in this day-and-age, that perhaps Charlie's > idea of having an extra of almost everything would warrant an > SLI/CrossFire?) solution. Specially since a cheapo extra would be a total > waste until you actually needed it. An SLI will be earning it's investment > while in use, AND having the beneficial ability of a dismantelment factor in > an emergency, to somewhat gloss over the smart-ass factor that inevatibly > comes with it as well? > > It's an interesting idea, though. The '9er' series should have the power we > want and a more finely tuned BIOS to fit the newly emerging technologies. > I'll go and see what Tom has to say about it (THG!). If they haven't spoken > allready, then the word must be practically on the threshold by now. Who > knows - in the final end collecting this kind of info could actually save > you a bit of money? > > > Tony. . . > > > > "Carlos" <Carlos@discussions.microsoft.com> wrote in message > news:159B5222-5C84-4861-8813-785F6C065B13@microsoft.com... > > Tony, > > The upcoming NVidia 9800 GX2 (dual-core GPU) might just be what you need. > > You will certainly need to shell out a lotta cash as opposed to "some" > cash, > > though! > > :) > > Carlos > > > > "Tony Sperling" wrote: > > > > > Thanks, John! Much as I thought. > > > > > > I could be wasting 'some' cash now, but there is also going to be 'some' > > > instant reward. Let's see, then, if the future brings more SLI or more > > > multi-core GPU's? > > > > > > > > > Tony. . . > > > > > > > > > > > > "John Barnes" <jbarnes@email.net> wrote in message > > > news:OBfdPi$fIHA.4744@TK2MSFTNGP06.phx.gbl... > > > > Not correct. nVidia whitepaper specifically states 'SLI is > completely > > > > implemented in NVIDIA's hardware and software, rendering the > technology > > > > transparent to affected applications, such as games.' > > > > > > > > Additionally states 'The graphics driver is effectively able to > analyze 3D > > > > scenes and divide the load sent to each video card according to the > fill > > > > rate, geometry, and shading demands with relatively little host > processor > > > > overhead. In some cases, you'll see the screen divided horizontally > into > > > > halves, but because the whole process is dynamic, workloads constantly > > > shift > > > > to reflect varying loads placed on each card. Thus, although one frame > > > might > > > > be cut 50/50, if the next is imbalanced, NVIDIA's driver will make > > > > adjustments to compensate.' > > > > > > > > > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > > > > news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > > > > > No. Each card uses its own memory. A program that is not SLI aware > will > > > > > just ignore the 2nd card altogether. > > > > > > > > > > Jay > > > > > > > > > > Tony Sperling wrote: > > > > >> Anybody know - does an application that is not SLI aware recognise > the > > > > >> combined Video Memory installed on the two cards? > > > > >> > > > > >> > > > > >> Tony. . . > > > > >> > > > > >> > > > > > > > > > > > >
Guest Carlos Posted March 7, 2008 Posted March 7, 2008 Re: OT. . . Tony, You can find useful comparison charts here: http://www.techarp.com/showarticle.aspx?artno=88&pgno=0 covering all the graphic cards (past, present, and future) Carlos "Tony Sperling" wrote: > Ha! > > No doubt. That IS the eternal direction 'evolution' has staked out for us, > mere mortals. Honestly, though, I don't know exactly what it is that I am > looking for, I was thinking, in this day-and-age, that perhaps Charlie's > idea of having an extra of almost everything would warrant an > SLI/CrossFire?) solution. Specially since a cheapo extra would be a total > waste until you actually needed it. An SLI will be earning it's investment > while in use, AND having the beneficial ability of a dismantelment factor in > an emergency, to somewhat gloss over the smart-ass factor that inevatibly > comes with it as well? > > It's an interesting idea, though. The '9er' series should have the power we > want and a more finely tuned BIOS to fit the newly emerging technologies. > I'll go and see what Tom has to say about it (THG!). If they haven't spoken > allready, then the word must be practically on the threshold by now. Who > knows - in the final end collecting this kind of info could actually save > you a bit of money? > > > Tony. . . > > > > "Carlos" <Carlos@discussions.microsoft.com> wrote in message > news:159B5222-5C84-4861-8813-785F6C065B13@microsoft.com... > > Tony, > > The upcoming NVidia 9800 GX2 (dual-core GPU) might just be what you need. > > You will certainly need to shell out a lotta cash as opposed to "some" > cash, > > though! > > :) > > Carlos > > > > "Tony Sperling" wrote: > > > > > Thanks, John! Much as I thought. > > > > > > I could be wasting 'some' cash now, but there is also going to be 'some' > > > instant reward. Let's see, then, if the future brings more SLI or more > > > multi-core GPU's? > > > > > > > > > Tony. . . > > > > > > > > > > > > "John Barnes" <jbarnes@email.net> wrote in message > > > news:OBfdPi$fIHA.4744@TK2MSFTNGP06.phx.gbl... > > > > Not correct. nVidia whitepaper specifically states 'SLI is > completely > > > > implemented in NVIDIA's hardware and software, rendering the > technology > > > > transparent to affected applications, such as games.' > > > > > > > > Additionally states 'The graphics driver is effectively able to > analyze 3D > > > > scenes and divide the load sent to each video card according to the > fill > > > > rate, geometry, and shading demands with relatively little host > processor > > > > overhead. In some cases, you'll see the screen divided horizontally > into > > > > halves, but because the whole process is dynamic, workloads constantly > > > shift > > > > to reflect varying loads placed on each card. Thus, although one frame > > > might > > > > be cut 50/50, if the next is imbalanced, NVIDIA's driver will make > > > > adjustments to compensate.' > > > > > > > > > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > > > > news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > > > > > No. Each card uses its own memory. A program that is not SLI aware > will > > > > > just ignore the 2nd card altogether. > > > > > > > > > > Jay > > > > > > > > > > Tony Sperling wrote: > > > > >> Anybody know - does an application that is not SLI aware recognise > the > > > > >> combined Video Memory installed on the two cards? > > > > >> > > > > >> > > > > >> Tony. . . > > > > >> > > > > >> > > > > > > > > > > > >
Guest jabloomf1230 Posted March 9, 2008 Posted March 9, 2008 Re: OT. . . And where does it say anything about the RAM on each video card? All that says is that SLI balances the load between the GPUs. If a program doesn't run in SLI mode, it only uses one GPU and the RAM on that card. John Barnes wrote: > Not correct. nVidia whitepaper specifically states 'SLI is completely > implemented in NVIDIA's hardware and software, rendering the technology > transparent to affected applications, such as games.' > > Additionally states 'The graphics driver is effectively able to analyze > 3D scenes and divide the load sent to each video card according to the > fill rate, geometry, and shading demands with relatively little host > processor overhead. In some cases, you'll see the screen divided > horizontally into halves, but because the whole process is dynamic, > workloads constantly shift to reflect varying loads placed on each card. > Thus, although one frame might be cut 50/50, if the next is imbalanced, > NVIDIA's driver will make adjustments to compensate.' > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... >> No. Each card uses its own memory. A program that is not SLI aware >> will just ignore the 2nd card altogether. >> >> Jay >> >> Tony Sperling wrote: >>> Anybody know - does an application that is not SLI aware recognise the >>> combined Video Memory installed on the two cards? >>> >>> >>> Tony. . . >>> >>>
Guest John Barnes Posted March 9, 2008 Posted March 9, 2008 Re: OT. . . May I suggest you look up the definition of 'transparent' as used in technology. Since the OP are technically savvy responding to unsourced posts is a waste of effort. You might consider studying up on how video cards, drivers and DirectX 10 interact. "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message news:usbuDPYgIHA.944@TK2MSFTNGP05.phx.gbl... > And where does it say anything about the RAM on each video card? All that > says is that SLI balances the load between the GPUs. If a program doesn't > run in SLI mode, it only uses one GPU and the RAM on that card. > > John Barnes wrote: >> Not correct. nVidia whitepaper specifically states 'SLI is completely >> implemented in NVIDIA's hardware and software, rendering the technology >> transparent to affected applications, such as games.' >> >> Additionally states 'The graphics driver is effectively able to analyze >> 3D scenes and divide the load sent to each video card according to the >> fill rate, geometry, and shading demands with relatively little host >> processor overhead. In some cases, you'll see the screen divided >> horizontally into halves, but because the whole process is dynamic, >> workloads constantly shift to reflect varying loads placed on each card. >> Thus, although one frame might be cut 50/50, if the next is imbalanced, >> NVIDIA's driver will make adjustments to compensate.' >> >> >> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message >> news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... >>> No. Each card uses its own memory. A program that is not SLI aware will >>> just ignore the 2nd card altogether. >>> >>> Jay >>> >>> Tony Sperling wrote: >>>> Anybody know - does an application that is not SLI aware recognise the >>>> combined Video Memory installed on the two cards? >>>> >>>> >>>> Tony. . . >>>> >>>>
Guest jabloomf1230 Posted March 9, 2008 Posted March 9, 2008 Re: OT. . . Why don't you post your thoughts on a video card website and see what kind of response that you get? SLI does not allow crossover use of RAM from one card to the other. For example, I have a PC with a 7950 GX2, which has two GPUs running in SLI. The card has 1 GB of RAM, but all the posted specs indicate clearly that each GPU has access to it own 512 MB of RAM. All SLI does is to either a) have the two GPUs render half of each frame or b) have the two GPUs render alternate frames. Scaling up tri- and Quad-SLI just splits things up in thirds or quarters. There is no magic to any of this and if a program is not SLI capable, it uses only the primary GPU. The other GPU does nothing unless you want it to drive a second monitor in multi-monitor mode. John Barnes wrote: > May I suggest you look up the definition of 'transparent' as used in > technology. Since the OP are technically savvy responding to unsourced > posts is a waste of effort. You might consider studying up on how video > cards, drivers and DirectX 10 interact. > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > news:usbuDPYgIHA.944@TK2MSFTNGP05.phx.gbl... >> And where does it say anything about the RAM on each video card? All >> that says is that SLI balances the load between the GPUs. If a program >> doesn't run in SLI mode, it only uses one GPU and the RAM on that card. >> >> John Barnes wrote: >>> Not correct. nVidia whitepaper specifically states 'SLI is >>> completely implemented in NVIDIA's hardware and software, rendering >>> the technology transparent to affected applications, such as games.' >>> >>> Additionally states 'The graphics driver is effectively able to >>> analyze 3D scenes and divide the load sent to each video card >>> according to the fill rate, geometry, and shading demands with >>> relatively little host processor overhead. In some cases, you'll see >>> the screen divided horizontally into halves, but because the whole >>> process is dynamic, workloads constantly shift to reflect varying >>> loads placed on each card. Thus, although one frame might be cut >>> 50/50, if the next is imbalanced, NVIDIA's driver will make >>> adjustments to compensate.' >>> >>> >>> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message >>> news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... >>>> No. Each card uses its own memory. A program that is not SLI aware >>>> will just ignore the 2nd card altogether. >>>> >>>> Jay >>>> >>>> Tony Sperling wrote: >>>>> Anybody know - does an application that is not SLI aware recognise the >>>>> combined Video Memory installed on the two cards? >>>>> >>>>> >>>>> Tony. . . >>>>> >>>>> >
Guest Tony Sperling Posted March 10, 2008 Posted March 10, 2008 Re: OT. . . If what the nVidia paper is saying, is correct (as I think we may well assume) does that not indicate that the video driver is in command? If, however, code has been written to implement it's own interpretation of what the SLI configuration should be doing, and since 'SLI awareness' definitely seems to be a factor (at least among gamers) - I think the driver will be allowing this. The Flightsim Developers have openly admited that none of their products as yet, are SLI aware. Nor are they coded to use multiple CPU's, which is one thing they are investigating the possibility of doing in the future! So, yes everybody seems to be more or less correct. As for the memory, I believe it would be strange if multiple cards, that are effectively buttoned up together, would be presenting the system with separate memory resources that would only be visible as a combined resource for code that was written in some specific way. 'Ignorant' and 'aware' code would logically be handled the same way as the driver's job is to do the graphics - not interpret the awareness factor of the code that wants to use it's resources? I thank everyone! I regard the question answered, and intend to read that paper closely (thanks, John!) If more questions arise, I may go and investigate on the nVidia forum - probably not a bad idea anyway? Jay: With two CPU's it would have to do that, they couldn't be competing for the same memory, could they? At least not if SLI mode can be switched off? On my 'Dual-Core' RAM too is devided among the two cores. . .if I'm not misstaken? Thanks! Tony. . . "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message news:%23l4ZpvjgIHA.2004@TK2MSFTNGP05.phx.gbl... > Why don't you post your thoughts on a video card website and see what > kind of response that you get? SLI does not allow crossover use of RAM > from one card to the other. For example, I have a PC with a 7950 GX2, > which has two GPUs running in SLI. The card has 1 GB of RAM, but all the > posted specs indicate clearly that each GPU has access to it own 512 MB > of RAM. > > All SLI does is to either a) have the two GPUs render half of each frame > or b) have the two GPUs render alternate frames. Scaling up tri- and > Quad-SLI just splits things up in thirds or quarters. There is no magic > to any of this and if a program is not SLI capable, it uses only the > primary GPU. The other GPU does nothing unless you want it to drive a > second monitor in multi-monitor mode. > > John Barnes wrote: > > May I suggest you look up the definition of 'transparent' as used in > > technology. Since the OP are technically savvy responding to unsourced > > posts is a waste of effort. You might consider studying up on how video > > cards, drivers and DirectX 10 interact. > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > > news:usbuDPYgIHA.944@TK2MSFTNGP05.phx.gbl... > >> And where does it say anything about the RAM on each video card? All > >> that says is that SLI balances the load between the GPUs. If a program > >> doesn't run in SLI mode, it only uses one GPU and the RAM on that card. > >> > >> John Barnes wrote: > >>> Not correct. nVidia whitepaper specifically states 'SLI is > >>> completely implemented in NVIDIA's hardware and software, rendering > >>> the technology transparent to affected applications, such as games.' > >>> > >>> Additionally states 'The graphics driver is effectively able to > >>> analyze 3D scenes and divide the load sent to each video card > >>> according to the fill rate, geometry, and shading demands with > >>> relatively little host processor overhead. In some cases, you'll see > >>> the screen divided horizontally into halves, but because the whole > >>> process is dynamic, workloads constantly shift to reflect varying > >>> loads placed on each card. Thus, although one frame might be cut > >>> 50/50, if the next is imbalanced, NVIDIA's driver will make > >>> adjustments to compensate.' > >>> > >>> > >>> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > >>> news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > >>>> No. Each card uses its own memory. A program that is not SLI aware > >>>> will just ignore the 2nd card altogether. > >>>> > >>>> Jay > >>>> > >>>> Tony Sperling wrote: > >>>>> Anybody know - does an application that is not SLI aware recognise the > >>>>> combined Video Memory installed on the two cards? > >>>>> > >>>>> > >>>>> Tony. . . > >>>>> > >>>>> > >
Guest Theo Posted March 10, 2008 Posted March 10, 2008 Re: OT. . . I think you need to research SLI a bit more. Just an extract: Both cards are given the same part of the 3D scene to render, but effectively half of the work load is sent to the slave card through a connector called the SLI Bridge. As an example, the master card works on the top half of the scene while the slave card works on the bottom half. When the slave card is done, it sends its output to the master card, which combines the two images to form one and then outputs the final render to the monitor. This does not agree with your hypothesis at all. jabloomf1230 wrote: > Why don't you post your thoughts on a video card website and see what > kind of response that you get? SLI does not allow crossover use of RAM > from one card to the other. For example, I have a PC with a 7950 GX2, > which has two GPUs running in SLI. The card has 1 GB of RAM, but all the > posted specs indicate clearly that each GPU has access to it own 512 MB > of RAM. > > All SLI does is to either a) have the two GPUs render half of each frame > or b) have the two GPUs render alternate frames. Scaling up tri- and > Quad-SLI just splits things up in thirds or quarters. There is no magic > to any of this and if a program is not SLI capable, it uses only the > primary GPU. The other GPU does nothing unless you want it to drive a > second monitor in multi-monitor mode. >
Guest jabloomf1230 Posted March 10, 2008 Posted March 10, 2008 Re: OT. . . Tony, CPUs are the beast of the OS and usage of RAM is up to Windows. Under SLI, the GPUs access to the video RAM is solely dependent on the video driver and from all public indications, the nVidia drivers don't allow the GPUs to use memory from another card under SLI. This may become more evident when the the 9800GX2 is released by nVidia in another week or so. This is another attempt by nVidia to sell a dual GPU card. The new video driver may allow such memory sharing for the 9800GX2, but we won't know for sure until the card and the new drivers that accompany the card are released. Jay Tony Sperling wrote: > If what the nVidia paper is saying, is correct (as I think we may well > assume) does that not indicate that the video driver is in command? > > If, however, code has been written to implement it's own interpretation of > what the SLI configuration should be doing, and since 'SLI awareness' > definitely seems to be a factor (at least among gamers) - I think the driver > will be allowing this. > > The Flightsim Developers have openly admited that none of their products as > yet, are SLI aware. Nor are they coded to use multiple CPU's, which is one > thing they are investigating the possibility of doing in the future! > > So, yes everybody seems to be more or less correct. As for the memory, I > believe it would be strange if multiple cards, that are effectively buttoned > up together, would be presenting the system with separate memory resources > that would only be visible as a combined resource for code that was written > in some specific way. > > 'Ignorant' and 'aware' code would logically be handled the same way as the > driver's job is to do the graphics - not interpret the awareness factor of > the code that wants to use it's resources? I thank everyone! I regard the > question answered, and intend to read that paper closely (thanks, John!) If > more questions arise, I may go and investigate on the nVidia forum - > probably not a bad idea anyway? > > Jay: > > With two CPU's it would have to do that, they couldn't be competing for the > same memory, could they? At least not if SLI mode can be switched off? On my > 'Dual-Core' RAM too is devided among the two cores. . .if I'm not misstaken? > > > Thanks! > > > Tony. . . > > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > news:%23l4ZpvjgIHA.2004@TK2MSFTNGP05.phx.gbl... >> Why don't you post your thoughts on a video card website and see what >> kind of response that you get? SLI does not allow crossover use of RAM >> from one card to the other. For example, I have a PC with a 7950 GX2, >> which has two GPUs running in SLI. The card has 1 GB of RAM, but all the >> posted specs indicate clearly that each GPU has access to it own 512 MB >> of RAM. >> >> All SLI does is to either a) have the two GPUs render half of each frame >> or b) have the two GPUs render alternate frames. Scaling up tri- and >> Quad-SLI just splits things up in thirds or quarters. There is no magic >> to any of this and if a program is not SLI capable, it uses only the >> primary GPU. The other GPU does nothing unless you want it to drive a >> second monitor in multi-monitor mode. >> >> John Barnes wrote: >>> May I suggest you look up the definition of 'transparent' as used in >>> technology. Since the OP are technically savvy responding to unsourced >>> posts is a waste of effort. You might consider studying up on how video >>> cards, drivers and DirectX 10 interact. >>> >>> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message >>> news:usbuDPYgIHA.944@TK2MSFTNGP05.phx.gbl... >>>> And where does it say anything about the RAM on each video card? All >>>> that says is that SLI balances the load between the GPUs. If a program >>>> doesn't run in SLI mode, it only uses one GPU and the RAM on that card. >>>> >>>> John Barnes wrote: >>>>> Not correct. nVidia whitepaper specifically states 'SLI is >>>>> completely implemented in NVIDIA's hardware and software, rendering >>>>> the technology transparent to affected applications, such as games.' >>>>> >>>>> Additionally states 'The graphics driver is effectively able to >>>>> analyze 3D scenes and divide the load sent to each video card >>>>> according to the fill rate, geometry, and shading demands with >>>>> relatively little host processor overhead. In some cases, you'll see >>>>> the screen divided horizontally into halves, but because the whole >>>>> process is dynamic, workloads constantly shift to reflect varying >>>>> loads placed on each card. Thus, although one frame might be cut >>>>> 50/50, if the next is imbalanced, NVIDIA's driver will make >>>>> adjustments to compensate.' >>>>> >>>>> >>>>> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message >>>>> news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... >>>>>> No. Each card uses its own memory. A program that is not SLI aware >>>>>> will just ignore the 2nd card altogether. >>>>>> >>>>>> Jay >>>>>> >>>>>> Tony Sperling wrote: >>>>>>> Anybody know - does an application that is not SLI aware recognise > the >>>>>>> combined Video Memory installed on the two cards? >>>>>>> >>>>>>> >>>>>>> Tony. . . >>>>>>> >>>>>>> > >
Guest jabloomf1230 Posted March 10, 2008 Posted March 10, 2008 Re: OT. . . Theo, I'm not sure what your point is? The original question was what happens when an SLI-enabled system encounters a program that is not SLI aware. Are you saying that in that circumstance, both GPUs are still used by the software to render graphics? Or that if only one GPU is use, it can use the RAM on the second card for loading textures, etc.? Jay Theo wrote: > I think you need to research SLI a bit more. Just an extract: > > Both cards are given the same part of the 3D scene to render, but > effectively half of the work load is sent to the slave card through a > connector called the SLI Bridge. As an example, the master card works on > the top half of the scene while the slave card works on the bottom half. > When the slave card is done, it sends its output to the master card, > which combines the two images to form one and then outputs the final > render to the monitor. > > This does not agree with your hypothesis at all. > > > jabloomf1230 wrote: >> Why don't you post your thoughts on a video card website and see what >> kind of response that you get? SLI does not allow crossover use of RAM >> from one card to the other. For example, I have a PC with a 7950 GX2, >> which has two GPUs running in SLI. The card has 1 GB of RAM, but all >> the posted specs indicate clearly that each GPU has access to it own >> 512 MB of RAM. >> >> All SLI does is to either a) have the two GPUs render half of each >> frame or b) have the two GPUs render alternate frames. Scaling up tri- >> and Quad-SLI just splits things up in thirds or quarters. There is no >> magic to any of this and if a program is not SLI capable, it uses only >> the primary GPU. The other GPU does nothing unless you want it to >> drive a second monitor in multi-monitor mode. >>
Guest Tony Sperling Posted March 11, 2008 Posted March 11, 2008 Re: OT. . . Thanks, I'll try and read up on the subject - it is 'clear' that things are getting 'foggy', but it's interesting all-the-same. Tony. . . "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message news:uPMZLzvgIHA.5900@TK2MSFTNGP02.phx.gbl... > Tony, > > CPUs are the beast of the OS and usage of RAM is up to Windows. Under > SLI, the GPUs access to the video RAM is solely dependent on the video > driver and from all public indications, the nVidia drivers don't allow > the GPUs to use memory from another card under SLI. This may become more > evident when the the 9800GX2 is released by nVidia in another week or > so. This is another attempt by nVidia to sell a dual GPU card. The new > video driver may allow such memory sharing for the 9800GX2, but we won't > know for sure until the card and the new drivers that accompany the card > are released. > > Jay > > Tony Sperling wrote: > > If what the nVidia paper is saying, is correct (as I think we may well > > assume) does that not indicate that the video driver is in command? > > > > If, however, code has been written to implement it's own interpretation of > > what the SLI configuration should be doing, and since 'SLI awareness' > > definitely seems to be a factor (at least among gamers) - I think the driver > > will be allowing this. > > > > The Flightsim Developers have openly admited that none of their products as > > yet, are SLI aware. Nor are they coded to use multiple CPU's, which is one > > thing they are investigating the possibility of doing in the future! > > > > So, yes everybody seems to be more or less correct. As for the memory, I > > believe it would be strange if multiple cards, that are effectively buttoned > > up together, would be presenting the system with separate memory resources > > that would only be visible as a combined resource for code that was written > > in some specific way. > > > > 'Ignorant' and 'aware' code would logically be handled the same way as the > > driver's job is to do the graphics - not interpret the awareness factor of > > the code that wants to use it's resources? I thank everyone! I regard the > > question answered, and intend to read that paper closely (thanks, John!) If > > more questions arise, I may go and investigate on the nVidia forum - > > probably not a bad idea anyway? > > > > Jay: > > > > With two CPU's it would have to do that, they couldn't be competing for the > > same memory, could they? At least not if SLI mode can be switched off? On my > > 'Dual-Core' RAM too is devided among the two cores. . .if I'm not misstaken? > > > > > > Thanks! > > > > > > Tony. . . > > > > > > > > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > > news:%23l4ZpvjgIHA.2004@TK2MSFTNGP05.phx.gbl... > >> Why don't you post your thoughts on a video card website and see what > >> kind of response that you get? SLI does not allow crossover use of RAM > >> from one card to the other. For example, I have a PC with a 7950 GX2, > >> which has two GPUs running in SLI. The card has 1 GB of RAM, but all the > >> posted specs indicate clearly that each GPU has access to it own 512 MB > >> of RAM. > >> > >> All SLI does is to either a) have the two GPUs render half of each frame > >> or b) have the two GPUs render alternate frames. Scaling up tri- and > >> Quad-SLI just splits things up in thirds or quarters. There is no magic > >> to any of this and if a program is not SLI capable, it uses only the > >> primary GPU. The other GPU does nothing unless you want it to drive a > >> second monitor in multi-monitor mode. > >> > >> John Barnes wrote: > >>> May I suggest you look up the definition of 'transparent' as used in > >>> technology. Since the OP are technically savvy responding to unsourced > >>> posts is a waste of effort. You might consider studying up on how video > >>> cards, drivers and DirectX 10 interact. > >>> > >>> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > >>> news:usbuDPYgIHA.944@TK2MSFTNGP05.phx.gbl... > >>>> And where does it say anything about the RAM on each video card? All > >>>> that says is that SLI balances the load between the GPUs. If a program > >>>> doesn't run in SLI mode, it only uses one GPU and the RAM on that card. > >>>> > >>>> John Barnes wrote: > >>>>> Not correct. nVidia whitepaper specifically states 'SLI is > >>>>> completely implemented in NVIDIA's hardware and software, rendering > >>>>> the technology transparent to affected applications, such as games.' > >>>>> > >>>>> Additionally states 'The graphics driver is effectively able to > >>>>> analyze 3D scenes and divide the load sent to each video card > >>>>> according to the fill rate, geometry, and shading demands with > >>>>> relatively little host processor overhead. In some cases, you'll see > >>>>> the screen divided horizontally into halves, but because the whole > >>>>> process is dynamic, workloads constantly shift to reflect varying > >>>>> loads placed on each card. Thus, although one frame might be cut > >>>>> 50/50, if the next is imbalanced, NVIDIA's driver will make > >>>>> adjustments to compensate.' > >>>>> > >>>>> > >>>>> "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > >>>>> news:eumhQT$fIHA.1188@TK2MSFTNGP04.phx.gbl... > >>>>>> No. Each card uses its own memory. A program that is not SLI aware > >>>>>> will just ignore the 2nd card altogether. > >>>>>> > >>>>>> Jay > >>>>>> > >>>>>> Tony Sperling wrote: > >>>>>>> Anybody know - does an application that is not SLI aware recognise > > the > >>>>>>> combined Video Memory installed on the two cards? > >>>>>>> > >>>>>>> > >>>>>>> Tony. . . > >>>>>>> > >>>>>>> > > > >
Guest Theo Posted March 11, 2008 Posted March 11, 2008 Re: OT. . . According to nVidia, and a few other sources, the fact that you have 2 graphics cards in a SLI configuration is totally transparent to the programs; i.e., they don't know, and don't care, whether there is one, two, or four video cards. The program is feeding the video information to the video system and that's all it know. The video software, along with the hardware, decide how to process the information. And, that's exactly what I quoted below: >> Both cards are given the same part of the 3D scene to render, but >> effectively half of the work load is sent to the slave card through a >> connector called the SLI Bridge. As an example, the master card works >> on the top half of the scene while the slave card works on the bottom >> half. When the slave card is done, it sends its output to the master >> card, which combines the two images to form one and then outputs the >> final render to the monitor. jabloomf1230 wrote: > Theo, > > I'm not sure what your point is? The original question was what happens > when an SLI-enabled system encounters a program that is not SLI aware. > Are you saying that in that circumstance, both GPUs are still used by > the software to render graphics? Or that if only one GPU is use, it can > use the RAM on the second card for loading textures, etc.? > > Jay > > Theo wrote: >> I think you need to research SLI a bit more. Just an extract: >> >> Both cards are given the same part of the 3D scene to render, but >> effectively half of the work load is sent to the slave card through a >> connector called the SLI Bridge. As an example, the master card works >> on the top half of the scene while the slave card works on the bottom >> half. When the slave card is done, it sends its output to the master >> card, which combines the two images to form one and then outputs the >> final render to the monitor. >> >> This does not agree with your hypothesis at all. >> >> >> jabloomf1230 wrote: >>> Why don't you post your thoughts on a video card website and see what >>> kind of response that you get? SLI does not allow crossover use of >>> RAM from one card to the other. For example, I have a PC with a 7950 >>> GX2, which has two GPUs running in SLI. The card has 1 GB of RAM, but >>> all the posted specs indicate clearly that each GPU has access to it >>> own 512 MB of RAM. >>> >>> All SLI does is to either a) have the two GPUs render half of each >>> frame or b) have the two GPUs render alternate frames. Scaling up >>> tri- and Quad-SLI just splits things up in thirds or quarters. There >>> is no magic to any of this and if a program is not SLI capable, it >>> uses only the primary GPU. The other GPU does nothing unless you want >>> it to drive a second monitor in multi-monitor mode. >>>
Guest Tony Sperling Posted March 11, 2008 Posted March 11, 2008 Re: OT. . . Conceptually, though, if software is polling for 'videoRamSize' will it report the combined RAM of both cards irregardless of the software being SLI aware, or not? It seems clear that the SLI enabled video driver will delegate the job the best it knows how, and the software wouldn't have to care - but if you want to know? I must confess, what I so naively thought was an innocent enough question has turned into something profundly philosophical. Doubt is entering the mind. I will try and investigate further. Tony. . . "Theo" <theo@discussions.microsoft.com> wrote in message news:%23wKbnZ2gIHA.5160@TK2MSFTNGP05.phx.gbl... > According to nVidia, and a few other sources, the fact that > you have 2 graphics cards in a SLI configuration is totally > transparent to the programs; i.e., they don't know, and > don't care, whether there is one, two, or four video cards. > The program is feeding the video information to the > video system and that's all it know. The video software, > along with the hardware, decide how to process the information. > > And, that's exactly what I quoted below: > > >> Both cards are given the same part of the 3D scene to > render, but > >> effectively half of the work load is sent to the slave > card through a > >> connector called the SLI Bridge. As an example, the > master card works > >> on the top half of the scene while the slave card works > on the bottom > >> half. When the slave card is done, it sends its output > to the master > >> card, which combines the two images to form one and then > outputs the > >> final render to the monitor. > > > jabloomf1230 wrote: > > Theo, > > > > I'm not sure what your point is? The original question was what happens > > when an SLI-enabled system encounters a program that is not SLI aware. > > Are you saying that in that circumstance, both GPUs are still used by > > the software to render graphics? Or that if only one GPU is use, it can > > use the RAM on the second card for loading textures, etc.? > > > > Jay > > > > Theo wrote: > >> I think you need to research SLI a bit more. Just an extract: > >> > >> Both cards are given the same part of the 3D scene to render, but > >> effectively half of the work load is sent to the slave card through a > >> connector called the SLI Bridge. As an example, the master card works > >> on the top half of the scene while the slave card works on the bottom > >> half. When the slave card is done, it sends its output to the master > >> card, which combines the two images to form one and then outputs the > >> final render to the monitor. > >> > >> This does not agree with your hypothesis at all. > >> > >> > >> jabloomf1230 wrote: > >>> Why don't you post your thoughts on a video card website and see what > >>> kind of response that you get? SLI does not allow crossover use of > >>> RAM from one card to the other. For example, I have a PC with a 7950 > >>> GX2, which has two GPUs running in SLI. The card has 1 GB of RAM, but > >>> all the posted specs indicate clearly that each GPU has access to it > >>> own 512 MB of RAM. > >>> > >>> All SLI does is to either a) have the two GPUs render half of each > >>> frame or b) have the two GPUs render alternate frames. Scaling up > >>> tri- and Quad-SLI just splits things up in thirds or quarters. There > >>> is no magic to any of this and if a program is not SLI capable, it > >>> uses only the primary GPU. The other GPU does nothing unless you want > >>> it to drive a second monitor in multi-monitor mode. > >>>
Guest jabloomf1230 Posted March 11, 2008 Posted March 11, 2008 Re: OT. . . As you know, Windows provides for the infamous >3 GB "memory hole" to accommodate the hardware memory address needs. For example, my 7950 GX2 uses one block of 512 MB of addresses for one GPU and a second distinct 512 MB block for the other GPU. Like I said before, I'm open to suggestions that somehow all the video RAM addresses are one big happy block, but everything from nVidia indicates otherwise. The drivers just aren't that sophisticated. Nor is the way that they handle SLI. I'm open to any suggestions, links, etc. indicating otherwise, but if you look at threads on this topic on such reputable graphics websites such as Guru of 3D and nVNews, you will see that there is certainty (as best as can be expected from a bunch of video gamers) as to this issue. The reason that this issue surfaced at all was that when the original 7900GX2 and 7950GX2 models came out, they were advertised with a (at that time) "whopping" 1GB of video RAM. This was subsequently proven to be "marchitecture", as nVidia was combining the video RAM for each GPU as if any one GPU could use all of it. The SLI solution chosen by nVidia is simple, if not elegant. You scale up by alternating frames or equal proportions of a frame and that's that. Even with that simple concept, Windows SLI drivers have been buggy from the beginning. And as you know, Vista didn't support SLI for a long time after its release, driving many potential customers away. As to the non-SLI issue, there is a wonderful little program called rthdribl, which has been used in the past to stress GPUs, especially those that are over clocked. Unfortunately it does not run in any SLI mode and in fact, if you try to run it in any SLI mode, it even runs slower than on one GPU. If you monitor GPU temperatures, while running rthdribl maxed out, you can see that the temperature of GPU0 will warm up nicely and the temperature of GPU1 will stay just above the idle value. This doesn't address the shared RAM issue, but it does demonstrate that non-SLI aware programs only rely on one GPU. Tony Sperling wrote: > Conceptually, though, if software is polling for 'videoRamSize' will it > report the combined RAM of both cards irregardless of the software being SLI > aware, or not? > > It seems clear that the SLI enabled video driver will delegate the job the > best it knows how, and the software wouldn't have to care - but if you want > to know? > > I must confess, what I so naively thought was an innocent enough question > has turned into something profundly philosophical. Doubt is entering the > mind. > > I will try and investigate further. > > > Tony. . . > > > "Theo" <theo@discussions.microsoft.com> wrote in message > news:%23wKbnZ2gIHA.5160@TK2MSFTNGP05.phx.gbl... >> According to nVidia, and a few other sources, the fact that >> you have 2 graphics cards in a SLI configuration is totally >> transparent to the programs; i.e., they don't know, and >> don't care, whether there is one, two, or four video cards. >> The program is feeding the video information to the >> video system and that's all it know. The video software, >> along with the hardware, decide how to process the information. >> >> And, that's exactly what I quoted below: >> >> >> Both cards are given the same part of the 3D scene to >> render, but >> >> effectively half of the work load is sent to the slave >> card through a >> >> connector called the SLI Bridge. As an example, the >> master card works >> >> on the top half of the scene while the slave card works >> on the bottom >> >> half. When the slave card is done, it sends its output >> to the master >> >> card, which combines the two images to form one and then >> outputs the >> >> final render to the monitor. >> >> >> jabloomf1230 wrote: >>> Theo, >>> >>> I'm not sure what your point is? The original question was what happens >>> when an SLI-enabled system encounters a program that is not SLI aware. >>> Are you saying that in that circumstance, both GPUs are still used by >>> the software to render graphics? Or that if only one GPU is use, it can >>> use the RAM on the second card for loading textures, etc.? >>> >>> Jay >>> >>> Theo wrote: >>>> I think you need to research SLI a bit more. Just an extract: >>>> >>>> Both cards are given the same part of the 3D scene to render, but >>>> effectively half of the work load is sent to the slave card through a >>>> connector called the SLI Bridge. As an example, the master card works >>>> on the top half of the scene while the slave card works on the bottom >>>> half. When the slave card is done, it sends its output to the master >>>> card, which combines the two images to form one and then outputs the >>>> final render to the monitor. >>>> >>>> This does not agree with your hypothesis at all. >>>> >>>> >>>> jabloomf1230 wrote: >>>>> Why don't you post your thoughts on a video card website and see what >>>>> kind of response that you get? SLI does not allow crossover use of >>>>> RAM from one card to the other. For example, I have a PC with a 7950 >>>>> GX2, which has two GPUs running in SLI. The card has 1 GB of RAM, but >>>>> all the posted specs indicate clearly that each GPU has access to it >>>>> own 512 MB of RAM. >>>>> >>>>> All SLI does is to either a) have the two GPUs render half of each >>>>> frame or b) have the two GPUs render alternate frames. Scaling up >>>>> tri- and Quad-SLI just splits things up in thirds or quarters. There >>>>> is no magic to any of this and if a program is not SLI capable, it >>>>> uses only the primary GPU. The other GPU does nothing unless you want >>>>> it to drive a second monitor in multi-monitor mode. >>>>> > >
Guest Tony Sperling Posted March 12, 2008 Posted March 12, 2008 Re: OT. . . Hmm. . .here is, finally one answer that seems to make sense to me: ***** SLI does not double the amount of video ram. Your available amount will be equal to the lowest amount on your cards. So if you put a 256mb card with a 512mb card, you would stil only have 256mb of video memory available. The reason for this is because each card renders a different frame, so they both have to have the exact same information in their memory to do the work. The information gets copied to each card, not doubled. This is a common SLI misconception. ***** The FAQ on the nVidia SLI Forum mentions the different schemes for dividing the display between the two cards, the one that divides the screen in two halves is said to be more computationally intensive because it must continuously calculate the spot where the division is applied (inserted - being imposed?). This would mean that the data being calculated on would have to be in memory on both cards at all times, I guess. Great! Each day, a new lesson learned! NB! - the FAQ is also mentioning that if you are not heavily into 'Gaming' and primarily using your machine for this purpose, SLI will not be economically attractive!!! (thanks, FAQ!) Tony. . . "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message news:%23K2EWx8gIHA.6032@TK2MSFTNGP03.phx.gbl... > As you know, Windows provides for the infamous >3 GB "memory hole" to > accommodate the hardware memory address needs. For example, my 7950 GX2 > uses one block of 512 MB of addresses for one GPU and a second distinct > 512 MB block for the other GPU. Like I said before, I'm open to > suggestions that somehow all the video RAM addresses are one big happy > block, but everything from nVidia indicates otherwise. The drivers just > aren't that sophisticated. Nor is the way that they handle SLI. I'm open > to any suggestions, links, etc. indicating otherwise, but if you look at > threads on this topic on such reputable graphics websites such as Guru > of 3D and nVNews, you will see that there is certainty (as best as can > be expected from a bunch of video gamers) as to this issue. > > The reason that this issue surfaced at all was that when the original > 7900GX2 and 7950GX2 models came out, they were advertised with a (at > that time) "whopping" 1GB of video RAM. This was subsequently proven to > be "marchitecture", as nVidia was combining the video RAM for each GPU > as if any one GPU could use all of it. The SLI solution chosen by nVidia > is simple, if not elegant. You scale up by alternating frames or equal > proportions of a frame and that's that. Even with that simple concept, > Windows SLI drivers have been buggy from the beginning. And as you know, > Vista didn't support SLI for a long time after its release, driving many > potential customers away. > > As to the non-SLI issue, there is a wonderful little program called > rthdribl, which has been used in the past to stress GPUs, especially > those that are over clocked. Unfortunately it does not run in any SLI > mode and in fact, if you try to run it in any SLI mode, it even runs > slower than on one GPU. If you monitor GPU temperatures, while running > rthdribl maxed out, you can see that the temperature of GPU0 will warm > up nicely and the temperature of GPU1 will stay just above the idle > value. This doesn't address the shared RAM issue, but it does > demonstrate that non-SLI aware programs only rely on one GPU. > > Tony Sperling wrote: > > Conceptually, though, if software is polling for 'videoRamSize' will it > > report the combined RAM of both cards irregardless of the software being SLI > > aware, or not? > > > > It seems clear that the SLI enabled video driver will delegate the job the > > best it knows how, and the software wouldn't have to care - but if you want > > to know? > > > > I must confess, what I so naively thought was an innocent enough question > > has turned into something profundly philosophical. Doubt is entering the > > mind. > > > > I will try and investigate further. > > > > > > Tony. . . > > > > > > "Theo" <theo@discussions.microsoft.com> wrote in message > > news:%23wKbnZ2gIHA.5160@TK2MSFTNGP05.phx.gbl... > >> According to nVidia, and a few other sources, the fact that > >> you have 2 graphics cards in a SLI configuration is totally > >> transparent to the programs; i.e., they don't know, and > >> don't care, whether there is one, two, or four video cards. > >> The program is feeding the video information to the > >> video system and that's all it know. The video software, > >> along with the hardware, decide how to process the information. > >> > >> And, that's exactly what I quoted below: > >> > >> >> Both cards are given the same part of the 3D scene to > >> render, but > >> >> effectively half of the work load is sent to the slave > >> card through a > >> >> connector called the SLI Bridge. As an example, the > >> master card works > >> >> on the top half of the scene while the slave card works > >> on the bottom > >> >> half. When the slave card is done, it sends its output > >> to the master > >> >> card, which combines the two images to form one and then > >> outputs the > >> >> final render to the monitor. > >> > >> > >> jabloomf1230 wrote: > >>> Theo, > >>> > >>> I'm not sure what your point is? The original question was what happens > >>> when an SLI-enabled system encounters a program that is not SLI aware. > >>> Are you saying that in that circumstance, both GPUs are still used by > >>> the software to render graphics? Or that if only one GPU is use, it can > >>> use the RAM on the second card for loading textures, etc.? > >>> > >>> Jay > >>> > >>> Theo wrote: > >>>> I think you need to research SLI a bit more. Just an extract: > >>>> > >>>> Both cards are given the same part of the 3D scene to render, but > >>>> effectively half of the work load is sent to the slave card through a > >>>> connector called the SLI Bridge. As an example, the master card works > >>>> on the top half of the scene while the slave card works on the bottom > >>>> half. When the slave card is done, it sends its output to the master > >>>> card, which combines the two images to form one and then outputs the > >>>> final render to the monitor. > >>>> > >>>> This does not agree with your hypothesis at all. > >>>> > >>>> > >>>> jabloomf1230 wrote: > >>>>> Why don't you post your thoughts on a video card website and see what > >>>>> kind of response that you get? SLI does not allow crossover use of > >>>>> RAM from one card to the other. For example, I have a PC with a 7950 > >>>>> GX2, which has two GPUs running in SLI. The card has 1 GB of RAM, but > >>>>> all the posted specs indicate clearly that each GPU has access to it > >>>>> own 512 MB of RAM. > >>>>> > >>>>> All SLI does is to either a) have the two GPUs render half of each > >>>>> frame or b) have the two GPUs render alternate frames. Scaling up > >>>>> tri- and Quad-SLI just splits things up in thirds or quarters. There > >>>>> is no magic to any of this and if a program is not SLI capable, it > >>>>> uses only the primary GPU. The other GPU does nothing unless you want > >>>>> it to drive a second monitor in multi-monitor mode. > >>>>> > > > >
Guest jabloomf1230 Posted March 12, 2008 Posted March 12, 2008 Re: OT. . . Yep. Mostly the video RAM is for loading textures, so each card needs its own RAM to do so. Check out this item at Fudzilla regarding the new dual GPU nVidia 9800GX2. The authors make a "snide" comment about the amount of video RAM: http://www.fudzilla.com/index.php?option=com_content&task=view&id=6251&Itemid=1 J Tony Sperling wrote: > Hmm. . .here is, finally one answer that seems to make sense to me: > > ***** > SLI does not double the amount of video ram. Your available amount will > be equal to the lowest amount on your cards. So if you put a 256mb card > with a 512mb card, you would stil only have 256mb of video memory > available. > > The reason for this is because each card renders a different frame, so > they > both have to have the exact same information in their memory to do the > work. The information gets copied to each card, not doubled. This is a > common SLI misconception. > ***** > > The FAQ on the nVidia SLI Forum mentions the different schemes for dividing > the display between the two cards, the one that divides the screen in two > halves is said to be more computationally intensive because it must > continuously calculate the spot where the division is applied (inserted - > being imposed?). This would mean that the data being calculated on would > have to be in memory on both cards at all times, I guess. Great! Each day, a > new lesson learned! > > NB! - the FAQ is also mentioning that if you are not heavily into 'Gaming' > and primarily using your machine for this purpose, SLI will not be > economically attractive!!! > > (thanks, FAQ!) > > > Tony. . . > > > "jabloomf1230" <jabloomf@nycap.rr.com> wrote in message > news:%23K2EWx8gIHA.6032@TK2MSFTNGP03.phx.gbl... >> As you know, Windows provides for the infamous >3 GB "memory hole" to >> accommodate the hardware memory address needs. For example, my 7950 GX2 >> uses one block of 512 MB of addresses for one GPU and a second distinct >> 512 MB block for the other GPU. Like I said before, I'm open to >> suggestions that somehow all the video RAM addresses are one big happy >> block, but everything from nVidia indicates otherwise. The drivers just >> aren't that sophisticated. Nor is the way that they handle SLI. I'm open >> to any suggestions, links, etc. indicating otherwise, but if you look at >> threads on this topic on such reputable graphics websites such as Guru >> of 3D and nVNews, you will see that there is certainty (as best as can >> be expected from a bunch of video gamers) as to this issue. >> >> The reason that this issue surfaced at all was that when the original >> 7900GX2 and 7950GX2 models came out, they were advertised with a (at >> that time) "whopping" 1GB of video RAM. This was subsequently proven to >> be "marchitecture", as nVidia was combining the video RAM for each GPU >> as if any one GPU could use all of it. The SLI solution chosen by nVidia >> is simple, if not elegant. You scale up by alternating frames or equal >> proportions of a frame and that's that. Even with that simple concept, >> Windows SLI drivers have been buggy from the beginning. And as you know, >> Vista didn't support SLI for a long time after its release, driving many >> potential customers away. >> >> As to the non-SLI issue, there is a wonderful little program called >> rthdribl, which has been used in the past to stress GPUs, especially >> those that are over clocked. Unfortunately it does not run in any SLI >> mode and in fact, if you try to run it in any SLI mode, it even runs >> slower than on one GPU. If you monitor GPU temperatures, while running >> rthdribl maxed out, you can see that the temperature of GPU0 will warm >> up nicely and the temperature of GPU1 will stay just above the idle >> value. This doesn't address the shared RAM issue, but it does >> demonstrate that non-SLI aware programs only rely on one GPU. >> >> Tony Sperling wrote: >>> Conceptually, though, if software is polling for 'videoRamSize' will it >>> report the combined RAM of both cards irregardless of the software being > SLI >>> aware, or not? >>> >>> It seems clear that the SLI enabled video driver will delegate the job > the >>> best it knows how, and the software wouldn't have to care - but if you > want >>> to know? >>> >>> I must confess, what I so naively thought was an innocent enough > question >>> has turned into something profundly philosophical. Doubt is entering the > >>> mind. >>> >>> I will try and investigate further. >>> >>> >>> Tony. . . >>> >>> >>> "Theo" <theo@discussions.microsoft.com> wrote in message >>> news:%23wKbnZ2gIHA.5160@TK2MSFTNGP05.phx.gbl... >>>> According to nVidia, and a few other sources, the fact that >>>> you have 2 graphics cards in a SLI configuration is totally >>>> transparent to the programs; i.e., they don't know, and >>>> don't care, whether there is one, two, or four video cards. >>>> The program is feeding the video information to the >>>> video system and that's all it know. The video software, >>>> along with the hardware, decide how to process the information. >>>> >>>> And, that's exactly what I quoted below: >>>> >>>> >> Both cards are given the same part of the 3D scene to >>>> render, but >>>> >> effectively half of the work load is sent to the slave >>>> card through a >>>> >> connector called the SLI Bridge. As an example, the >>>> master card works >>>> >> on the top half of the scene while the slave card works >>>> on the bottom >>>> >> half. When the slave card is done, it sends its output >>>> to the master >>>> >> card, which combines the two images to form one and then >>>> outputs the >>>> >> final render to the monitor. >>>> >>>> >>>> jabloomf1230 wrote: >>>>> Theo, >>>>> >>>>> I'm not sure what your point is? The original question was what > happens >>>>> when an SLI-enabled system encounters a program that is not SLI aware. >>>>> Are you saying that in that circumstance, both GPUs are still used by >>>>> the software to render graphics? Or that if only one GPU is use, it > can >>>>> use the RAM on the second card for loading textures, etc.? >>>>> >>>>> Jay >>>>> >>>>> Theo wrote: >>>>>> I think you need to research SLI a bit more. Just an extract: >>>>>> >>>>>> Both cards are given the same part of the 3D scene to render, but >>>>>> effectively half of the work load is sent to the slave card through a >>>>>> connector called the SLI Bridge. As an example, the master card works >>>>>> on the top half of the scene while the slave card works on the bottom >>>>>> half. When the slave card is done, it sends its output to the master >>>>>> card, which combines the two images to form one and then outputs the >>>>>> final render to the monitor. >>>>>> >>>>>> This does not agree with your hypothesis at all. >>>>>> >>>>>> >>>>>> jabloomf1230 wrote: >>>>>>> Why don't you post your thoughts on a video card website and see > what >>>>>>> kind of response that you get? SLI does not allow crossover use of >>>>>>> RAM from one card to the other. For example, I have a PC with a 7950 >>>>>>> GX2, which has two GPUs running in SLI. The card has 1 GB of RAM, > but >>>>>>> all the posted specs indicate clearly that each GPU has access to it >>>>>>> own 512 MB of RAM. >>>>>>> >>>>>>> All SLI does is to either a) have the two GPUs render half of each >>>>>>> frame or b) have the two GPUs render alternate frames. Scaling up >>>>>>> tri- and Quad-SLI just splits things up in thirds or quarters. There >>>>>>> is no magic to any of this and if a program is not SLI capable, it >>>>>>> uses only the primary GPU. The other GPU does nothing unless you > want >>>>>>> it to drive a second monitor in multi-monitor mode. >>>>>>> >>> > >
Recommended Posts