discussion SKYMTL, can you please comment???

[MAC]

If they were lying that much about this then ATI/AMD would step in to stop it. Putting words in a company's mouth is probably a very, very bad idea.

A) AMD has very little pull with most websites (that they don't support).

B) It's called satire, it's protected.

 

[AkG]

Probably falls under the parody / satire rules of the Fair Use Laws..and it IS so blatant that no one should take it seriously....and yes Malaysia has freedom of speech (WHOIS states it is reg'ed in Selangor, MY). As long as it is not seditious Malaysians are free to lampoon whomever they want.

Good idea to peeve off ATI?....depends on how big a sense of humor they have. But no, prolly not a good idea. But that hardly stops anyone these days. Look at how many sites were breaking NDA over the NV 480.

Edit NINJA'ed by MAC!

 

[Sushi Warrior]

ATI has only 1 tesselator unit, I think if they turned it off it just wouldn't work.

 

[SKYMTL]

Patently false unfortunately.

The ATI HD 5000 series uses pretty much the same architecture that has been in place since the HD 2900 series with some modifications made along the way. As such, it has a single fixed function stage tessellator so it is either "enabled" or "disabled" but as you probably already guessed, it was never disabled...even in past series where it uses were limited.

 

[Delavan]

Thanks for debunking this myth/urban legend/crock!
Sounded too good to be true.
Surprising that some techsites are spreading this stuff. Well, tells us a lot about the so called "tech sites"....
Thanks SKY!

 

[LarkStarr]

from the article :

<table align="center" border="1" bordercolor="#000000" cellpadding="4" cellspacing="0" width="550"><tbody><tr bgcolor="#333333"><td bordercolor="#000000" align="center" bgcolor="#ffcc00" width="90">Date</td> <td bordercolor="#000000" align="center" bgcolor="#ffcc00" width="60">Revision</td> <td bordercolor="#000000" align="center" bgcolor="#ffcc00" width="400">Revision History</td> </tr> <tr> <td bordercolor="#000000" align="center" bgcolor="#f9f9f9">01-04-2010
</td> <td bordercolor="#000000" align="center" bgcolor="#f9f9f9">1.0
</td> <td bordercolor="#000000" bgcolor="#f9f9f9">Initial Release.
</td></tr></tbody></table>

just sayin'

 

[CMetaphor]

Now maybe AMD should answer by actually enabling Sideport on their dual-gpu cards, lol. That'll show these guys :bleh:.

 

[Zero82z]

Now maybe AMD should answer by actually enabling Sideport on their dual-gpu cards, lol. That'll show these guys :bleh:.

The reason they left it disabled is because it didn't provide any performance benefit and consumed more power. I'm sure they would enable it if they could actually get tangible performance increases from doing so.

 

[CMetaphor]

The reason they left it disabled is because it didn't provide any performance benefit and consumed more power. I'm sure they would enable it if they could actually get tangible performance increases from doing so.

I don't see how eliminating an in-between pci-express controller to communicate between the two cores could possibly be a bad thing... or consume more power? If anything, one less chip means less power, less latency? Possible increase in peformance, less complicated PCB design?

 

[Zero82z]

I don't see how eliminating an in-between pci-express controller to communicate between the two cores could possibly be a bad thing... or consume more power? If anything, one less chip means less power, less latency? Possible increase in peformance, less complicated PCB design?

Sideport isn't a replacement for the PLX chip, it's an addition to it. I remember that when the 4870X2 came out, there was talk about AMD enabling Sideport soon to improve performance, but a little while after they basically said that they weren't going to bother because it didn't help.