R520 informanina, part II (E3, etc)

digitalwanderer said:
DaveBaumann said:
http://www.beyond3d.com/news/images/20050519_r520.jpg
:oops:

OMG, you daring tease....thanks Dave!!! :D

Heh, I wonder how many harddrives that sucker is on already. :D

Now, was that the only picture he took, the most detailed, or the least detailed? :LOL:

Let me save Dave and Digi the trouble: "Yes".
 
geo said:
R300 was released a bit before DX9, wasn't it?

Yes it was, but I don't think those two cases are comparable.

Tho, speaking for myself, I haven't divined yet whether the WGF2.0 transition is so extreme from a performance profile pov, as to make wanting a base of Longhorn users out there in advance highly desirable before release. In other words, even if nonWGF2.0 cards can co-exist with Longhorn (obviously they can, hence WGF1.0), and WGF2.0 cards can co-exist with WinXP (almost certainly they will), whether the performance for gaming will suck so heartily when "fish out of water" that not too many gamers would be willing to do it other than very short term.

Example: a hypothetical R520 summer 2005 and an early 2006 R580 would fit better in line with a mid 2006 R600 or a let's say late 2006/early 2007 release?

Besides if someone would tell me that WGF2.0 GPUs have been layed out for 90nm, then anything would be possible. With 65nm though the story would be different.....
 
digitalwanderer said:
rwolf said:
I am surpised that they haven't licensed Sappires liquid metal cooler.
The IHVs always seem to be one step behind the aftermarket coolers, methinks it's a matter of focus.

It's also one of the few places that the partners can still differentiate themselves from the reference design of these extremely complex cards. They probably encourage ATI to at least leave them a few things to add value to the product.
 
It's hard to tell from the picture Dave posted, but unless that Intel motherboard is micro-ATX, it would seem as though the R520 PCB is currently longer than previous ATI high-end boards (X850 XTPE). If it is longer, what could account for the increased PCB real-estate, since there should be no Rialto bridge chip on that particular card, due to it being shown in an Intel PCI-E system?

Cheers,


BrynS
 
IgnorancePersonified said:
I think it is a small mobo... the fan ontop of the cpu fan off centre... is that for the north bridge?
I thought that at first too, but it kind of looks like it might actually be in the case window and not inside the case.
 
tEd said:
You would assume that if the chips get smaller the boards would too. :rolleyes:

How do you come up with that? The amount of electronics around the chip isn't going to become any less.
 
_xxx_ said:
tEd said:
You would assume that if the chips get smaller the boards would too. :rolleyes:

How do you come up with that? The amount of electronics around the chip isn't going to become any less.

It was more of an ironic comment. As technologie evolves(especially with electronic devices) it usually gets smaller and smaller but still graphiccards end up bigger and bigger no matter what.
 
A bit of a side question, but that Prey game that Anand mentions in the article linked to in the first post... is it... you know... Prey? You know, as in the Prey from ages ago that never was released?
Or did someone just take the name?
 
horvendile said:
A bit of a side question, but that Prey game that Anand mentions in the article linked to in the first post... is it... you know... Prey? You know, as in the Prey from ages ago that never was released?
Or did someone just take the name?

There's several recent threads in the gaming forum. It's not the same game not made by the same people (though licenced out by 3DR), but it does take the name, and quite a few of the same concepts. It's been made with the Doom3 engine, rather than the original idea of the Prey engine to drive the Prey game.
 
BrynS said:
It's hard to tell from the picture Dave posted, but unless that Intel motherboard is micro-ATX, it would seem as though the R520 PCB is currently longer than previous ATI high-end boards (X850 XTPE). If it is longer, what could account for the increased PCB real-estate, since there should be no Rialto bridge chip on that particular card, due to it being shown in an Intel PCI-E system?

Cheers,


BrynS

Think memory...
 
Back
Top