My notes from Las Vegas Launch Event
- Thread starter DemoCoder
- Start date
megadrive0088
Regular
Exactly. I'm more interested in when NV35 and R350 are coming, and what they'll have over NV30 and R300. I wonder what the NV34 is, that was also annouced to have taped out!
I'm expecting to be somewhat disappointed with NV30, but usually, Nvidia refreshes are OUTSTANDING. I'm hoping Nvidia can bounce back quickly with an un-delayed refresh (not a speed bump) with NV35 (or NV34????)
I'd like nothing more than to see R350 vs NV35 (or NV34) in the spring.
The ATI fanboys on this forum are very annoying and extremely vocal.
'desperate to catch ATI, thus had to clock it higher',
'performance numbers are FUD',
'6* ATI AA = 8* NV30 AA'
'R350 will be out by february'
'Less bandwidth = slower performance in reality'
'crappy anisotropic quality'
'In reality, will only ship by may'
etc etc
Guys. We essentially don't know ANYTHING yet. While all of the above maybe true, they might be 100% wrong!
Wait for the reviews to hit before passing assumptions off as reality.
'desperate to catch ATI, thus had to clock it higher',
'performance numbers are FUD',
'6* ATI AA = 8* NV30 AA'
'R350 will be out by february'
'Less bandwidth = slower performance in reality'
'crappy anisotropic quality'
'In reality, will only ship by may'
etc etc
Guys. We essentially don't know ANYTHING yet. While all of the above maybe true, they might be 100% wrong!
Wait for the reviews to hit before passing assumptions off as reality.
Joe DeFuria
Legend
Ummm...Fred...
First of all, considering this is a LAUNCH, the fact that "we don't know anything yet" should raise some flags....
Who is trying to pass anything off like you mentioned as "reality." All we can do is speculate based on the information we have...which as I said is surprisingly little for a launch...
First of all, considering this is a LAUNCH, the fact that "we don't know anything yet" should raise some flags....
Who is trying to pass anything off like you mentioned as "reality." All we can do is speculate based on the information we have...which as I said is surprisingly little for a launch...
Both camps have been extremely vocal of late.
megadrive0088
Regular
the R350 by Feb might really be true.
at least by spring, it should be out. even then, R350 wouldn't be out long after NV30
at least by spring, it should be out. even then, R350 wouldn't be out long after NV30
TheMightyPuck
Newcomer
Fred said:
One wouldn't have to be an ATI f@nboi to make any of those comments except maybe the anisotropic QUALITY remark.
Dreadnought
Newcomer
DaveBaumann said:
You'll do yourself a nasty injury on that fense trying to be so neutral
DemoCoder said:
DC - did you think the DOF effects looked kinda pixellated?
Randell said:
I'm not Demo, but I'll throw in my opinion as well. Being a 3D artist myself, I'll also try to judge them as any pre-rendered art I've seen, as the convergence is now obvious.
- The Yeah demo uses art created by Spellcraft Studios, credits should only go for Nvidia's programmers; and IMHO the lack of good lighting ruins the quality of it quite a lot. Just throwing in a second, non-shadowing and tiny bit colored light would do wonders to it.
- The fairy's head is a modified version of Aki's mesh from Final Fantasy IMHO. The shape of the noseis the same, the eyes look pretty similar and the texture is also quite familiar. I believe that either Nvidia has hired an ex-Square artist (the movie studio has closed some months ago) or they've used and modified the assets that they've recieved for the GF3 Final Fantasy techdemo.
The shader seems to be an improved version though, as the surface parameters are dependent on the angle relative to the camera (which is more natural). However, notice that it lacks bump mapping
All in all, it is the most impressive one, and if you wouldn't tell someone that it's realtime, it could go as a quite good piece of prerendered art.
- The old car looks decent, but the lighting could be better. The chrome parts should use Fresnel reflections, they look unrealistic (and if they already do, they should be fine-tuned). Aliasing is very disturbing throughout the edges of the car.
- The DOF demo has horrible lighting, objects and texturing. No offense to those who made it, but it looks like typical beginner's 3D art. The aliasing is disturbing here as well.
So, I think that the techdemo team itself does not look like they're that more talented than the guys at ATi - they have managed to get their hands on some nice source material though. It is a practice that other IHVs should start to follow as soon as possible, as inhouse teams will quickly get too small to produce prerendered quality content...
Has there been any word of multi-chip/card configurations?
Also, I think that some previews said that the chip was going to be in a flip-chip package, like the R300. Is this correct? It's just that in all the pictures of the card, you cannot see the top of the core like you can on the R300. It appears to have a silver covering on top. Is this to further aid cooling?
Also, I think that some previews said that the chip was going to be in a flip-chip package, like the R300. Is this correct? It's just that in all the pictures of the card, you cannot see the top of the core like you can on the R300. It appears to have a silver covering on top. Is this to further aid cooling?
martrox said:
You were expecting maybe....plastic, wood...steel?.....
This is what has been known for decades in the industry as a "Product Announcement"....It announces a new and upcoming product and supplies a targeted shipping date. When a product begins shipping we call that a "Product Release."....
(Just kidding...the whole "paper launch" thing really bugs me.)
DemoCoder said:
I would think that not only is nVidia waiting to establish final silicon, chip yields, etc., but that there might be a wait on the 500MHz DDRII ram as well.
The UT2K3 numbers really puzzle me. The game is one of the most cpu limited I've ever seen. What are they running the nv30 on--Hammer? Or did the "accidentally" compare a fly-by to a bot match...
Silent running must mean the thing is dang loud when it's running...
nv30 must be hot as a firecracker at stock speeds.I expected no less, but nice to hear it confirmed and hopefully that will put the rumor mill to bed.
How do you know that wasn't the UT2K3 engine? It'll do softshadows and ought to handle "DoomIII-ish" 3D with ease.
Shades of 3dfx--think the feature will at last gain some respect?
WaltC said:The UT2K3 numbers really puzzle me. The game is one of the most cpu limited I've ever seen. What are they running the nv30 on--Hammer? Or did the "accidentally" compare a fly-by to a bot match...
The UK-Gamer site has the same numbers as DC posted. However, his numbers are photos of the slides presented and in the lower right corner of the shots are some system specs. The D3, UT2K3, Nature, and Q3 runs are reported as "3GHz P4, 512 MB RAM, WinXP." The SpecView Pref numbers where run on a "3GHz P4, ???? RAM, Granite Bay, Win2K SP3."
Link:
http://www.ukgamer.com/article.php4?id=208&page=3
Edit: Sorry about the double post. Took me too long to type the message.
Yes, the depth of field demo had some horrible pixelation artifacts when it was maximally out of focus. I chalk it up to bad coding. I've seen a similar effect done with bog standard DX8 that looked better.
I was also unimpressed by the rusted truck, the lighting looked too artificial, but then again, doing realistic diffuse lighting is.
I thought "Dawn" and "Ogre" were cool. However, while Orge was artistically fairly cool, I still have to give the edge to ATI's Animusic Demo, and moreover, their Debevec and Two-Tone demos for showing how higher precision can make a difference. (the split screen effects are nice)
And to answer the comments about the Sweeney demo, it was NOT the UT2k3 engine. It was stated to be a research project and it was more than just softshadows, it was a unified lighting model like Doom3 will full 100% shadowing everywhere (including softshadows) and also lots of other volumetric lighting effects. Sweeney said the main character that was walking around the screen (a Knight in armor with a torch) had 1+million polygons and that the NV30 could comfortably render about 20 of these at once. However, I doubt the full resolution mesh is actually being used.
And to answer your question about the benchmarks of Unreal, et al, they had them running on OEM stations outside the main conference, and you could play around with all the NV30 demos. However I was not able to get a look at benchmark numbers because the benchmark demos were in a special room for VIPs (probably for OEMs and important ISVs like EA, Epic, etc).
I must emphasize again, they had real, live, running hardware that anyone could play on, presumably running @ 500Mhz (dont know about memory) so none of the demos were canned, faked, etc.
I was told by several NVidia and OEM people I could buy one in January, so believe it or not for what it's worth.
The beginning the presentation almost made me faint: GeForce + 3dfx animation was shown combining into GeforceFx? Was the NV30 a Gigapixel tiler!? HOLYCOW! What a surprise. Reality turned out to be far more mundane, but I was still happy that they managed to get it running at 500Mhz with 1Ghz RAM. Think about it: 4gigapixels/s, possiblty 4-8 giga-shader-ops per second (depending on the dispatch rate), and 6-8X FSAA? That's potentially 32-64giga-ops/sec.
I am waiting with caution to see how many other features are present: displacement maps, gamma correct AA, Video features, dual output?, and how does LMA3 work, etc.
My overall feeling is: Thank gawd it's not a huge disappointment. It is performance competitive with ATI, so we will soon have two high performance DX9 chips on the market, and hopefully, ATI and NVidia can together force everyone to upgrade over time, and create a large platform for developers to take advantage of these features.
I was also unimpressed by the rusted truck, the lighting looked too artificial, but then again, doing realistic diffuse lighting is.
I thought "Dawn" and "Ogre" were cool. However, while Orge was artistically fairly cool, I still have to give the edge to ATI's Animusic Demo, and moreover, their Debevec and Two-Tone demos for showing how higher precision can make a difference. (the split screen effects are nice)
And to answer the comments about the Sweeney demo, it was NOT the UT2k3 engine. It was stated to be a research project and it was more than just softshadows, it was a unified lighting model like Doom3 will full 100% shadowing everywhere (including softshadows) and also lots of other volumetric lighting effects. Sweeney said the main character that was walking around the screen (a Knight in armor with a torch) had 1+million polygons and that the NV30 could comfortably render about 20 of these at once. However, I doubt the full resolution mesh is actually being used.
And to answer your question about the benchmarks of Unreal, et al, they had them running on OEM stations outside the main conference, and you could play around with all the NV30 demos. However I was not able to get a look at benchmark numbers because the benchmark demos were in a special room for VIPs (probably for OEMs and important ISVs like EA, Epic, etc).
I must emphasize again, they had real, live, running hardware that anyone could play on, presumably running @ 500Mhz (dont know about memory) so none of the demos were canned, faked, etc.
I was told by several NVidia and OEM people I could buy one in January, so believe it or not for what it's worth.
The beginning the presentation almost made me faint: GeForce + 3dfx animation was shown combining into GeforceFx? Was the NV30 a Gigapixel tiler!? HOLYCOW! What a surprise. Reality turned out to be far more mundane, but I was still happy that they managed to get it running at 500Mhz with 1Ghz RAM. Think about it: 4gigapixels/s, possiblty 4-8 giga-shader-ops per second (depending on the dispatch rate), and 6-8X FSAA? That's potentially 32-64giga-ops/sec.
I am waiting with caution to see how many other features are present: displacement maps, gamma correct AA, Video features, dual output?, and how does LMA3 work, etc.
My overall feeling is: Thank gawd it's not a huge disappointment. It is performance competitive with ATI, so we will soon have two high performance DX9 chips on the market, and hopefully, ATI and NVidia can together force everyone to upgrade over time, and create a large platform for developers to take advantage of these features.