NVIDIA G92 : Pre-review bits and pieces

Arun said:
* Triangle setup is also completely unmodified, with 0.5 tri/clk setup, 1 tri/clk culling and only 1 vertex/clk output to the post-T&L FIFO (without attributes, even in Z-only passes).
Click to expand...
OK, I'm really confused now, I thought G80 was 1 triangle per clock setup rate?!!!

Jawed
 
Jawed said:
OK, I'm really confused now, I thought G80 was 1 triangle per clock setup rate?!!!
Click to expand...
I thought that too at one point, so I can't blame you - my own test, as well as others' (including 3DMark06's) are very clear however: triangle setup is 0.5/clk. The triangles getting rejected based on depth tests don't seem to be an exception either...

If anyone knows of a case where you can get up to 1 tri/clk outside of backface culling, please let me know. I certainly cannot seem to find one, however! As for how the different chips on the market compare (speculated for unannounced products; note that I haven't been able to test most of those with my own tool):

R600/RV670/RV630: 1 tri/clk.
RV610: 0.5 tri/clk (iirc).

G92/G80/G84: 0.5 tri/clk.
G86/G98: 0.25 tri/clk (iirc).
 
Arun said:
Extra tidbits:

* NVIDIA confirmed to us that the chip supports GDDR4, but they obviously wouldn't comment on unannounced products.
* Blending is still half-rate, unlike R6xx which is full-rate (more important now given the number of ROPs). While you'd be bandwidth limited anyway for FP16 blending, it is clear that you can be ROP-limited for INT8/FP10 blending; not a major bottleneck, but still noteworthy.
* Triangle setup is also completely unmodified, with 0.5 tri/clk setup, 1 tri/clk culling and only 1 vertex/clk output to the post-T&L FIFO (without attributes, even in Z-only passes). Every single one of those metrics is at least 50% lower than R600's, and this is most likely a very real bottleneck in certain cases.
Click to expand...

It doesn't matter what specs are LESS than the R600's because in real world performance its faster than the R600 and the G80 8800GTS! Speed is what counts and my lowly 2900XT loses to this. Now as far as the new RV670 and R680's due out shortly, thats another story. Hopefully they'll be strong performers. As for now, its useless to say the 8800GT has lesser specs than anything out now because in the end it is MUCH FASTER. However, if I'm spending my hard earned cash I would like some DX10.1 implementation just for the hell of it. New tech should always have the newest features.
 
Rys said:
8800 GT Board Thoughts

* NVIDIA thermal solution engineers need to take off the earmuffs and nuke the wax buildup. I hate saying that.
Click to expand...

It's quite loud at boot-up, but in normal (3D-)operation we've measured no more than 0,8 sone, which is reasonably quiet.

This did, in fact, surprise me as well since i was willing to bet on a loud cooling solution when i first saw the tiny fan.
 
Quasar said:
It's quite loud at boot-up, but in normal (3D-)operation we've measured no more than 0,8 sone, which is reasonably quiet.

This did, in fact, surprise me as well since i was willing to bet on a loud cooling solution when i first saw the tiny fan.
Click to expand...

What would classify as a "normal" 3d operation? Light gaming?
 
Quasar said:
This did, in fact, surprise me as well since i was willing to bet on a loud cooling solution when i first saw the tiny fan.
Click to expand...

It's thin, but not tiny. :smile:

IMG_9918.jpg
 
Well, they do get points for single slot in my book, even if it comes with a temporarily annoying loud boot-up whir.
 
Geo said:
Well, they do get points for single slot in my book, even if it comes with a temporarily annoying loud boot-up whir.
Click to expand...
It's like three seconds. I think you're making a bigger deal of it than it really is...
 
Tim Murray said:
NV claimed it was better than EATM. It's certainly better than the old TRMS, but I haven't seen EATM to be able to really compare. If somebody wants a HL2 Ep1 saved game I used to make screenshots of the new TRMS mode to compare against EATM, I'd be happy to give it to you.
Click to expand...

World of WarCraft was a great test subject for "EATM" and was very welcomed. Too bad it was tak'n out.

Example1:
SSAA:
http://img514.imageshack.us/img514/296/wowssaavh1.jpg

EATM:
http://img453.imageshack.us/img453/6603/woweatmzt7.jpg

Example 2:
SSAA:
http://img441.imageshack.us/img441/746/westfallssaauu1.jpg

EATM:
http://img266.imageshack.us/img266/8849/westfalleatmho5.jpg
 
Tim Murray said:
It's like three seconds. I think you're making a bigger deal of it than it really is...
Click to expand...

Hey, I owned an X1900XT. I know from boot-up whirs. :p I didn't intend to make a big deal out of it in the sense of suggesting it was a reason for someone to not buy the card. I'd easily trade getting single-slot for 3 seconds of boot-up whir. Having said that, I don't see a rational reason why the boot-up whir really should be there in these cases when it ought to be fixable (whether it be X1900XT or 8800GT). Picking at nits? Youbetcha. That's the way progress is made.
 
You must log in or register to reply here.

Similar threads

T
GPU Stability testing for Nvidia cards - PowerSweep
Replies
0
Views
445
T2098
T
Cyan
Epic gives the go-ahead and you can now legally download Unreal Gold and Unreal Tournament 99 for free, two PC classics. Epic doesn't care anymore?
Replies
3
Views
90
Davros
D
Z
Some PS5Pro and Resident Evil 9 thoughts
2
Replies
21
Views
1K
Zombietoid
Z
D
Speculation and Rumors: Nvidia Blackwell ...
45 46 47
Replies
928
Views
103K
Kaotik
Kaotik
D
RTX Video HDR modded to work in games, RTX Game HDR?
Replies
7
Views
2K
DegustatoR
D
Back
Top