Most Significant Graphics Technology for 2005
- Thread starter rwolf
- Start date
IgnorancePersonified
Regular
A 512bit memeory bus. Don't know why but I think it will yield more for less R&D. Doesn't really constitute "significant technological advance"....
As a left over xmas wish I'd like to see less horseshit surrounding GFX chip/card releases.
As a left over xmas wish I'd like to see less horseshit surrounding GFX chip/card releases.
This is just a guess... feel free to disagree and enlighten
But my guess would be whatever is in the Xbox 2, which is rumored to be released at the end of 2005, will be the most "significant technological advance in computer graphics technology". Why? While the Xbox2 is not a PC necessarily, it will do a couple things that affect PCs:
1) Beyond developers trying to keep the lowest common denominator in view, I think the consoles tend to hold the PC market back some in regards to games' graphics. On the flip side I also think when new consoles are released it is a new "minimum" standard that is completed. Based on the outdated Xbox2 specs, it looks like the new consoles will rival high end video cards. My guess would be over the next 24 months we see more and more PC games shifting away from DX7 as a baseline (e.g. look how D3 and HL2 run fairly well on older HW) and begin hitting DX9 as the basic featureset.
2) The Xbox2 will be introducing a unified shader architecture. How this affects the PC market (good or bad), I do not know, but with WGF and R600 coming down the pipe and Nvidia looking at a non-unified structure from the HW point of view, this will definately be a significant advance. I am curious to see how this type of architecture affects content creation, because theoretically being able to dedicated all your Vertex Shader tasks should help out 3D workstation apps.
3) If the rumors are correct, there will be a "XBox2 PC". Imagine millions of PCs with high-mid range video cards in it. That is a nice lowest common denominator! Forget FX 5200s and Radeon 9200s!
Just my opinion... but then again R520 and NV50 could totally blow us away, or something totally different could. Who knows... shall be exciting!
But my guess would be whatever is in the Xbox 2, which is rumored to be released at the end of 2005, will be the most "significant technological advance in computer graphics technology". Why? While the Xbox2 is not a PC necessarily, it will do a couple things that affect PCs:
1) Beyond developers trying to keep the lowest common denominator in view, I think the consoles tend to hold the PC market back some in regards to games' graphics. On the flip side I also think when new consoles are released it is a new "minimum" standard that is completed. Based on the outdated Xbox2 specs, it looks like the new consoles will rival high end video cards. My guess would be over the next 24 months we see more and more PC games shifting away from DX7 as a baseline (e.g. look how D3 and HL2 run fairly well on older HW) and begin hitting DX9 as the basic featureset.
2) The Xbox2 will be introducing a unified shader architecture. How this affects the PC market (good or bad), I do not know, but with WGF and R600 coming down the pipe and Nvidia looking at a non-unified structure from the HW point of view, this will definately be a significant advance. I am curious to see how this type of architecture affects content creation, because theoretically being able to dedicated all your Vertex Shader tasks should help out 3D workstation apps.
3) If the rumors are correct, there will be a "XBox2 PC". Imagine millions of PCs with high-mid range video cards in it. That is a nice lowest common denominator! Forget FX 5200s and Radeon 9200s!
Just my opinion... but then again R520 and NV50 could totally blow us away, or something totally different could. Who knows... shall be exciting!
IgnorancePersonified
Regular
Are unified shaders going to be available in the PC space in 2005?
be good to see the base video card being of dx9 or above feature sets!
hmmmm: get this error today when posting "Could not connect to smtp host : 111 : Connection refused"
be good to see the base video card being of dx9 or above feature sets!
hmmmm: get this error today when posting "Could not connect to smtp host : 111 : Connection refused"
IgnorancePersonified
Regular
True. I guess i'm skeptical as to wether it will be introduced or not. Seems a pretty busy (new)year of releases already what with all the catching up to do on last years product releases. For MS it's of the upmost importance to get r500 out the door while i don't see the pressure on ati.
Not necessarily the quality of the render per-se, but the realism of the render. Confused yet? Multicore CPUs will allow for better simulation in realtime, so the quality of the render may not go up, but the realism of simulated animations (not pre-canned... pre canned = bad) that go along with those renders will.
Then you have a more direct impact that multicore CPUs could have. If anything can be offloaded to the CPU instead of choking the GPU needlessly with it then you may directly improve the quality of the image being rendered. Maybe if developers had the option of handing off (generic example that's heard over and over again) to a beefy CPU to free up the GPU they would.
Lesson one kids.... never type while drunk.
Happy Easter, and cheers, good health.
Newyears
Later
Then you have a more direct impact that multicore CPUs could have. If anything can be offloaded to the CPU instead of choking the GPU needlessly with it then you may directly improve the quality of the image being rendered. Maybe if developers had the option of handing off (generic example that's heard over and over again) to a beefy CPU to free up the GPU they would.
Lesson one kids.... never type while drunk.
Happy Easter, and cheers, good health.
Newyears
Later
Yeah. The best we can hope for in new 3D technology this year would be multisampling FSAA for FP16 rendertargets, and (though much, much less likely) a higher-precision buffer format for the target of tonemapping for FP16 rendering (i.e. 10-10-10-2).ANova said:
Dual-core CPU's, if they do arrive this year, will be high-end toys and won't have much impact. Multicore CPU's won't be interesting until developers get used to splitting up tasks into multiple threads, and CPU vendors start shipping 4-core+ CPU's.
Edit: Though some low-end SM3 parts would be a significant step forward, too. I think we can expect these by about the middle of the year from nVidia.
Most Significant Graphics Technology for 2005: 90nm!
It's not so much a specific technology IMHO, but rather the fact that the two top IHVs have such tight competition this year.
THAT is the most signifigant 3D-thingy for me this year, and I feel it's just been a win/win year for the consumer. 8)
THAT is the most signifigant 3D-thingy for me this year, and I feel it's just been a win/win year for the consumer. 8)
Joe DeFuria
Legend
Chalnoth said:
SM3 in the low-end means absolutely nothing in the grand scheme of things IMO.
The low end inflection point is "base" DX9 functionality. (SM 2.0). When nVidia and ATI move SM 3.0 into the low end is about as technologically important a gargle blaster.
I think every major CPU company is making them for 2k5, and I don't yet see why they'd be delayed. And they should be released all over the spectrum, so I don't think they'll be high-end toys.Chalnoth said:
The X2 is likely to have at least one dual-core processor, and I don't see how $300 consoles are considered high-end toys.
Would that be a regular gargle blaster or a pan-galactic gargle blaster? :|Joe DeFuria said:
digitalwanderer said:
apart from the fact that both IHV's bleeding-edge parts were not available in any volume, and still aren't if we're realistic now. The 9800 XT was far more-readily available than either bleeding-edge part this time around. :?
Sure there are two very good products in the X800 Pro and the 6800GT that have been fairly readily available, but there hasn't been realistic volume for either X800 XT Platinum Edition, or 6800Ultra IMHO. At least not on the global scale of yesteryear's bleeding edge.
I keep hearing that, but I still also keep remembering how after only about 20 minutes of comparing I was able to order & get delivered an X800 pro VIVO I wanted within 42 hours.bigz said:
I know the real XT PEs are rare, but it wasn't hard to get a hold of a card and make your own.
