nvidia just can't stop

W

whql

Regular
So, evidently they weren't satisfied with the x800 pdf, they are now bs'ing about ati's pci express solution:

http://www.amdreview.com/forums/showthread.php?s=&threadid=15474

Looks like they are still touting that x-ray around and saying that there's an intenal bridge solution - really, who'd fall for this, it would just be more expensive for ati to produce since it would need two bus interfaces in the silicon.

Ati has been talking about the number of oem slots they are getting with pci express - looks like nvidia are just being a bunch of sore loosers.
 
I got the FUD bit, but I don't understand why.

PCI-e isn't out, isn't near coming out, and when it does it's going to be adopted to a whole lot slower than everyone seems to think it is.

I guess I don't understand the big need to be spreading FUD about it when it isn't here yet nor really due any time soon, as well as just drawing more attention to the fact that you just can't get anything other than ATi cards if you want a next gen card currently. :(
 
digitalwanderer said:
I got the FUD bit, but I don't understand why.

PCI-e isn't out, isn't near coming out, and when it does it's going to be adopted to a whole lot slower than everyone seems to think it is.

I guess I don't understand the big need to be spreading FUD about it when it isn't here yet nor really due any time soon, as well as just drawing more attention to the fact that you just can't get anything other than ATi cards if you want a next gen card currently. :(
Click to expand...

Oh that - that's just Nvidia's PR machine out of control as usual. Damage control in the midst of panic.
 
arent nvidia using a bridge chip on their supposedly "native" PCI-E solution as well (this time around they moved it to the gpu package)
 
gokickrocks said:
arent nvidia using a bridge chip on their supposedly "native" PCI-E solution as well (this time around they moved it to the gpu package)
Click to expand...
They were bragging about it a few months ago... :?
 
digitalwanderer said:
I got the FUD bit, but I don't understand why.

PCI-e isn't out, isn't near coming out, and when it does it's going to be adopted to a whole lot slower than everyone seems to think it is.

I guess I don't understand the big need to be spreading FUD about it when it isn't here yet nor really due any time soon, as well as just drawing more attention to the fact that you just can't get anything other than ATi cards if you want a next gen card currently. :(
Click to expand...

I think you're wrong.

Intel is the dominant chipset on the market, once they start shipping pci-e (and I believe they intend to ship only pci-e) it will become widespread very rapidly.

I bet major OEM's don't give a crap about nv40 and r420 at this point, it won't impact their sales in the least.
 
SsP45 said:
weeds said:
http://www.theinquirer.net/?article=16613
Click to expand...

That's different. That's bridging a PCI Express chip to use AGP.
Click to expand...

And sadly it seems as if the inq is trying to play it off as ATI is being hypocritical. From my understanding a PCI-E to AGP bridge will perform nearly identical to a native AGP implementation, but if you go the other way around you lose performance.

Could someone with a little more knowledge back me up? (Or shoot me down either ones good)
 
Killer-Kris said:
SsP45 said:
weeds said:
http://www.theinquirer.net/?article=16613
Click to expand...

That's different. That's bridging a PCI Express chip to use AGP.
Click to expand...

And sadly it seems as if the inq is trying to play it off as ATI is being hypocritical. From my understanding a PCI-E to AGP bridge will perform nearly identical to a native AGP implementation, but if you go the other way around you lose performance.

Could someone with a little more knowledge back me up? (Or shoot me down either ones good)
Click to expand...

Well we don't really know if one implementation will be faster than the other, but generally you would think that it would be less problematic to make the faster interface run slower than the other way around.
 
AlphaWolf said:
Well we don't really know if one implementation will be faster than the other, but generally you would think that it would be less problematic to make the faster interface run slower than the other way around.
Click to expand...
I think you underestimate the power of PR to affect that equation. ;)
 
Somebody said:
NVIDIA bashes ATI through “ATI’s PCI Express to AGP Bridge is Limited to AGP 8X speed and cannot run at Grantsdale or Alderwood’s Maximum Bandwidthâ€￾
Click to expand...

Not sure I even understand this. A PCIe-to-AGP bridge means that on PCIe it doesn't need, or use, a bridge. The bridge is employed in the case of an AGP x8 bus environment instead of PCIe. What are they [it,them] saying, exactly?
 
WaltC said:
Somebody said:
NVIDIA bashes ATI through “ATI’s PCI Express to AGP Bridge is Limited to AGP 8X speed and cannot run at Grantsdale or Alderwood’s Maximum Bandwidthâ€￾
Click to expand...

Not sure I even understand this. A PCIe-to-AGP bridge means that on PCIe it doesn't need, or use, a bridge. The bridge is employed in the case of an AGP x8 bus environment instead of PCIe. What are they [it,them] saying, exactly?
Click to expand...

That future chips won't support AGP and that Ati will use a bridge on AGP-boards. This should not be surprise to anybody as it does not make sense to keep making new native-AGP chips.

The PCIe boards will off cause not need a bridge.
 
You must log in or register to reply here.

Similar threads

G
Apple Thunderbolt Display Mini-review (Grall definition of "mini"...)
Replies
12
Views
9K
Mendel
M
A
NVIDIA Chipset Discussion (MCP55, Nehalem & Misc.)
Replies
6
Views
4K
INKster
I
A
NVIDIA's Record Q407: ~$1.2B Revenue/~$300M Net Income
Replies
18
Views
7K
Sxotty
S
M
Nvidia Ageia acquisition - console edition
Replies
17
Views
5K
sevanig
S
K
ATI Soundview conference
Replies
5
Views
4K
kemosabe
K
Back
Top